WO2023284447A1 - 云边协同的数据传输方法、服务器及存储介质 - Google Patents

云边协同的数据传输方法、服务器及存储介质 Download PDF

Info

Publication number
WO2023284447A1
WO2023284447A1 PCT/CN2022/097602 CN2022097602W WO2023284447A1 WO 2023284447 A1 WO2023284447 A1 WO 2023284447A1 CN 2022097602 W CN2022097602 W CN 2022097602W WO 2023284447 A1 WO2023284447 A1 WO 2023284447A1
Authority
WO
WIPO (PCT)
Prior art keywords
transmission
data
communication links
communication
cloud server
Prior art date
Application number
PCT/CN2022/097602
Other languages
English (en)
French (fr)
Inventor
陈共龙
Original Assignee
腾讯科技(深圳)有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from CN202110793799.4A external-priority patent/CN113489643B/zh
Application filed by 腾讯科技(深圳)有限公司 filed Critical 腾讯科技(深圳)有限公司
Publication of WO2023284447A1 publication Critical patent/WO2023284447A1/zh

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/24Multipath
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L45/00Routing or path finding of packets in data switching networks
    • H04L45/12Shortest path evaluation
    • H04L45/124Shortest path evaluation using a combination of metrics
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/01Protocols
    • H04L67/10Protocols in which an application is distributed across nodes in the network

Definitions

  • the present application relates to the field of Internet technology, specifically to the field of cloud technology, and in particular to a cloud-edge collaborative data transmission method, device, server, and storage medium.
  • Cloud-edge collaboration refers to the close collaboration between edge computing and cloud computing.
  • Edge computing is performed by edge servers
  • cloud computing is performed by cloud servers.
  • edge servers can receive signaling data sent by the user equipment, and forward the received signaling data to a corresponding cloud server, so that the cloud server can perform subsequent processing according to the signaling data.
  • how to transmit signaling data between edge servers and cloud servers has become a research hotspot.
  • An embodiment of the present application provides a cloud-side collaborative data transmission method, the data transmission method includes:
  • link planning is performed between the edge server and the target cloud server to obtain multiple communication links
  • the signaling data is transmitted through the K communication links.
  • An embodiment of the present application provides a cloud-side collaborative data transmission method, the data transmission method includes:
  • K is a positive integer; the K communication links are selected from multiple communication links according to the network quality optimization goal; the multiple communication links The path is obtained by performing link planning between the edge server and the target cloud server according to the data transmission strategy between the edge server and the target cloud server, and the target cloud server refers to the signaling data The cloud server to be reached;
  • the current cloud server refers to: through the K communication links and the edge server Cloud server for communication;
  • An embodiment of the present application provides a data transmission device for cloud-edge collaboration, and the data transmission device includes:
  • a determining unit configured to determine a target cloud server where the signaling data to be transmitted needs to arrive;
  • a processing unit configured to perform link planning between the edge server and the target cloud server according to a data transmission strategy between the edge server and the target cloud server, to obtain multiple communication links;
  • the processing unit is further configured to select K communication links satisfying the network quality optimization goal from the plurality of communication links according to the network quality optimization goal, where K is a positive integer;
  • a transmission unit configured to transmit the signaling data through the K communication links.
  • An embodiment of the present application provides a data transmission device for cloud-edge collaboration, and the data transmission device includes:
  • the transmission unit is used to receive the signaling data transmitted by the edge server through K communication links, K is a positive integer; the K communication links are selected from multiple communication links according to the network quality optimization goal; the The plurality of communication links are obtained by link planning between the edge server and the target cloud server according to the data transmission strategy between the edge server and the target cloud server, and the target cloud server is The cloud server to which the signaling data needs to arrive;
  • a processing unit configured to perform data processing on the received signaling data if the destination address of the received signaling data points to the current cloud server, the current cloud server refers to: through the K communication links A cloud server communicating with the edge server;
  • the transmission unit is further configured to forward the received signaling data to the target cloud server through an intranet if the received signaling data is not directed to the current cloud server.
  • An embodiment of the present application provides a server, the server includes an input interface and an output interface, and the server further includes: a processor adapted to implement one or more computer-readable instructions; and a computer storage medium, the computer The storage medium stores one or more computer-readable instructions; the one or more instructions are suitable for being loaded by the processor and executing the above cloud-edge collaborative data transmission method.
  • An embodiment of the present application provides a non-volatile computer-readable storage medium, the non-volatile computer-readable storage medium stores one or more computer-readable instructions, and the one or more computer-readable instructions
  • the read instruction is suitable for being loaded by the processor and executed by the above cloud-edge collaborative data transmission method.
  • FIG. 1a is a system architecture diagram of a cloud-edge collaborative data transmission system provided by an embodiment of the present application
  • Fig. 1b is a schematic diagram of a cloud-edge collaborative data transmission solution provided by an embodiment of the present application
  • FIG. 2 is a schematic flow diagram of a cloud-edge collaborative data transmission method provided by an embodiment of the present application
  • FIG. 3a is a schematic diagram of signaling data transmission through K communication links provided by an embodiment of the present application.
  • Fig. 3b is another schematic diagram of signaling data transmission through K communication links provided by the embodiment of the present application.
  • FIG. 4 is a schematic flowchart of a cloud-side collaborative data transmission method provided by another embodiment of the present application.
  • FIG. 5 is a schematic flowchart of a cloud-side collaborative data transmission method provided by another embodiment of the present application.
  • Fig. 6a is a schematic diagram of a transmission cost optimization solution and a network quality optimization goal provided by an embodiment of the present application
  • FIG. 6b is a schematic flow diagram of a cloud server performing related operations on received data packets provided by an embodiment of the present application
  • FIG. 7 is a schematic structural diagram of a data transmission device for cloud-edge collaboration provided by an embodiment of the present application.
  • FIG. 8 is a schematic structural diagram of a cloud-edge collaborative data transmission device provided by another embodiment of the present application.
  • FIG. 9 is a schematic structural diagram of a server provided by an embodiment of the present application.
  • the embodiment of this application proposes a cloud-side collaborative data transmission scheme based on multi-channel optimization.
  • the data transmission scheme can adaptively select an appropriate data transmission strategy according to the network quality optimization target and the network quality of the communication link ( It can be referred to simply as the transmission strategy) and transmission mode, and flexibly select the communication link with better network quality for data transmission, so as to meet the needs of users while reducing the cost of users and improving the flexibility of data transmission, thereby improving user Experience and user stickiness.
  • the embodiment of the present application proposes a cloud-edge collaborative data transmission system; as shown in Figure 1a, the data transmission system may include but not limited to: edge servers (or called edge nodes), and cloud servers (or cloud nodes) deployed in different regions.
  • the cloud servers can communicate through the intranet (or private network, local area network, etc.);
  • the edge server can be in the public network (or external network, wide area network, Internet (Internet)
  • the operator establishes one or more operator links with each cloud server, and communicates with the corresponding cloud server based on the established operator links.
  • the edge server may include a link quality monitoring layer, a multiplex optimization layer, and a transport layer;
  • the link quality monitoring layer is mainly responsible for collecting the network quality of each operator link between the edge server and each cloud server, In order to facilitate the subsequent arrangement and analysis of data transmission strategies and transmission methods;
  • the multi-channel optimization layer is mainly responsible for arrangement and analysis of appropriate data transmission strategies, and according to the network quality optimization goals, comprehensively consider the various operators collected by the link quality monitoring layer The network quality of the link and the communication cost of each operator link, select the appropriate transmission mode and the better operator link;
  • the transport layer is mainly responsible for the data transmission compiled and analyzed according to the multi-channel optimization layer The operator's link selected by the policy, transmission mode, and multiplex optimization layer is used for data transmission.
  • Fig. 1a is only an exemplary representation of the system architecture of the data transmission system, and is not limited thereto.
  • Figure 1a only exemplarily represents an edge server, but in practical applications, the number of edge servers included in the data transmission system is not limited to 1, and may also be 2, 3 or even more;
  • Figure 1a only exemplarily characterizes two cloud servers located in Region 1 and Region 2, but in actual applications, the data transmission system may also include more cloud servers; for another example, the Both the edge server and one of the cloud servers are located in region 1, but in practical applications, the edge server and each cloud server may be located in different regions, and so on.
  • any of the above-mentioned servers can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or it can provide Cloud service, cloud database, cloud computing, cloud function, cloud storage, network service, cloud communication, middleware service, domain name service, security service, CDN (Content Delivery Network, content distribution network), and big data and artificial intelligence platforms, etc.
  • Cloud servers for basic cloud computing services etc.
  • any server may be located outside the blockchain network or within the blockchain network, which is not limited.
  • the models of the edge server and the cloud server can be common models, and the edge server and the cloud server can be equipped with network cards with a specified network speed (such as 100Gbps (1000 megabits per second)) to increase bandwidth.
  • n public network communication addresses can be configured for the edge server, and m public network communication addresses can be configured for each cloud server; that is, n public network communication addresses of the edge server and each The m public network communication addresses of the cloud server can be provided by one or more operators.
  • n and m are both positive integers, and the values of n and m can be the same or different; the so-called public network communication address refers to the communication address used in the public network, which can be, for example, a public network IP (Internet Protocol , the protocol for interconnection between networks) addresses.
  • a public network communication address in the edge server and a public network communication address in the cloud server can be used to establish an operator link; based on this, the transport layer in the edge server can use the n public network communication addresses of the edge server and There are m public network communication addresses of each cloud server, and n ⁇ m operator links are established between the edge server and each cloud server. It should be noted that when the public network communication address is a public network IP address, the operator link may also be called an IP link.
  • the link quality monitoring layer in the edge server can detect the network quality of each operator link in the transport layer. Specifically, the link quality monitoring layer may first collect the data transmission status of each operator link in the transport layer, and the data transmission status of each operator link may include but not limited to: Whether there is packet loss in the transmitted data, and when there is packet loss, the number of lost data packets, etc., whether there is delay, and if there is delay, the specific delay length and other delay conditions.
  • the link quality monitoring layer can determine the network quality of each operator link according to the collected data transmission conditions of each operator link; Two link quality monitoring arrays of length W are maintained between each cloud server, and the two link quality monitoring arrays between the edge server and any cloud server can be used for storage: between the edge server and any cloud server The delay time of each operator link and the packet loss rate of each communication link.
  • the letter a is used to indicate the edge server
  • the letter x is used to indicate any cloud server
  • the delay D i_ax [W] in seconds
  • the link quality monitoring layer can regularly (that is, every time T, T is the preset duration) detect the network quality of each operator link in the transport layer, so as to realize monitoring every time T according to Each link quality monitoring array maintained is updated according to the received network quality, and then each link quality monitoring array is fed back to the multi-path optimization layer every time T.
  • the link quality monitoring layer may also detect the network quality in real time, or perform the network quality detection only once, which is not limited.
  • the multi-channel optimization layer in the edge server can determine the cloud server (subsequently referred to as the target cloud server) that the signaling data needs to reach, and determine the distance between the edge server and the target cloud server. Data transfer policy.
  • link planning can be performed between the edge server and the target cloud server to obtain multiple communication links, which are essentially operator links. Then, the planned network quality of each communication link can be determined based on each link quality monitoring array fed back by the link quality monitoring layer; Based on factors such as the public network communication cost of each communication link, plan and formulate a lower-cost transmission method that can meet user needs.
  • the transmission method may be a method of selecting a single communication link for data transmission (abbreviated as a single-link transmission method), or may be a method of selecting at least two communication links for data transmission (referred to as a multi-link transmission method). method), without limitation.
  • the transmission mode is a multi-link transmission mode
  • at least two selected communication links can be used to perform multi-link concurrent transmission of signaling data, or to perform multi-link redundant transmission of signaling data;
  • concurrent transmission means that the data transmitted by each communication link is different, and the redundant transmission means that the data transmitted by each communication link is the same or partly the same.
  • the multi-channel selection layer can send the determined data transmission strategy and transmission method to the transport layer, so that the transmission layer can transmit the signaling data according to the data transmission strategy and transmission method , to transmit the signaling data to the target cloud server.
  • the data transmission scheme proposed in the embodiment of the present application can establish a multiple Links of operators, and enable cloud access and cloud intranet intercommunication functions, dynamically plan multiple communication links, so as to further combine network quality optimization goals, taking into account factors such as network quality of communication links and public network communication costs , flexibly select a suitable communication link from the planned multiple communication links for data transmission, so that not only the communication link with better network quality can be used for data transmission, but also the reliability of data transmission can be improved, thereby improving The effect of data transmission can also effectively reduce user costs.
  • the embodiment of the present application proposes a cloud-edge collaborative data transmission method.
  • the cloud-edge collaborative data transmission method can be executed by the aforementioned edge server, and the edge server can communicate with multiple cloud servers.
  • the data transmission method for cloud-side collaboration may include the following steps S201-S204:
  • the edge server after the edge server obtains the signaling data to be transmitted, it can first parse the signaling data to obtain the destination address of the signaling data.
  • the destination address can be any cloud server among multiple cloud servers.
  • the public network communication address of the server then, the edge server can determine the cloud server indicated by the destination address of the signaling data as the target cloud server that the signaling data needs to reach, that is, the target cloud server refers to the target cloud server that the signaling data needs to reach cloud server.
  • the signaling data may be any type of data.
  • the signaling data may be system operation data generated during the operation of the edge server or the user equipment.
  • the signaling data may be user operation data generated according to the user's operation behavior on the client when the user is using the client.
  • the client here can be an audio and video client, a game client, a browser client, etc.
  • the signaling data can be user operation data generated according to the user's audio and video playback operation on the audio and video client, or according to the user's User operation data generated for the game operation of the game client, or user operation data generated according to the user's information selection operation/information browsing operation/information search operation on the browser client, etc., etc.
  • the games supported by the above-mentioned game client can be cloud games or ordinary games, which is not limited; the so-called cloud games can also be called game on demand, which is a game method based on cloud computing.
  • Cloud computing is an Internet-based computing method; in the cloud game scenario, the game is not run on the game client of the player user, but in the cloud game server; the cloud game server encodes the game screen involved in the game scene It is a video stream, which is transmitted to the game client of the player user through the network for playback; correspondingly, an ordinary game refers to a game that is directly run in the game client of the player user.
  • S202 Perform link planning between the edge server and the target cloud server according to the data transmission strategy between the edge server and the target cloud server to obtain multiple communication links.
  • multiple transmission strategies can be supported between the edge server and the target cloud server; specifically, when the edge server and the target cloud server are located in the same region, the edge server can be supported to directly transfer data to Transmission to the target cloud server; when the edge server and the target cloud server are located in different regions, it can still support the edge server to directly transmit data to the target cloud server through the public network, or support the edge server to transmit data to the target cloud server through other cloud servers
  • the target cloud server for example, can support the edge server to connect to a nearby cloud server, and the nearby connected cloud server transmits data to the target cloud server.
  • the so-called “nearest access” means: according to the distance between the edge server and each remote server, select the nearest cloud server for access, so as to perform data transmission with the nearest remote server;
  • the distance can refer to: the geographical distance between the geographical location of the edge server and the remote server, or can refer to: the communication distance between the public network communication address of the edge server and the public network communication address of the cloud server, There is no limit to this.
  • the edge server can select one of the above-mentioned multiple transmission strategies according to the actual situation as the data transmission strategy between the edge server and the target cloud server; that is, the data transmission
  • the policy can be used to indicate: the edge server transmits data to the target cloud server through the public network, or the data transmission policy can be used to indicate: the edge server connects to the cloud server nearby, and the cloud server connected nearby transmits the data to the target cloud server through the intranet target cloud server.
  • the two situations of the data transmission strategy are listed here as examples, and are not exhaustive.
  • the data transmission policy can be used to indicate: the edge server randomly accesses a cloud server, and the randomly accessed cloud server transmits data to the target cloud server through the intranet; as another example, in other embodiments
  • the data transmission strategy can be used to indicate that the edge server accesses the cloud server with the best performance according to the performance of each cloud server, and the connected cloud server transmits data to the target cloud server through the intranet, and so on.
  • the edge server transmits data to the target cloud server through the public network; Links, as multiple communication links.
  • the specific implementation method of step S202 can be: the edge server and the edge The multiple operator links between the cloud servers connected to the server nearby serve as multiple communication links.
  • the edge server is configured with n public network communication addresses
  • the cloud server is configured with m public network communication addresses, where n and m are both positive integers
  • an operator link uses a public network communication address of the edge server, and A public network communication address of the cloud server is established.
  • step S202 when the data transmission policy is used to indicate that the edge server randomly accesses a cloud server, or accesses information such as a cloud server with the best performance, the specific implementation of step S202 is the same as that used when accessing the nearest cloud server. The specific implementation manners are similar and will not be repeated here.
  • the edge server can obtain the network quality of each of the planned multiple communication links; Select K communication links that meet the network quality optimization goal; K is a positive integer, and the value of K is less than or equal to the number of planned communication links.
  • the network quality optimization target may include at least one of the following: a packet loss rate to be achieved, a delay time to be achieved, and the like.
  • the network quality of each communication link may include at least one of the following: a packet loss rate of the communication link, a delay time of the communication link, and the like.
  • the edge server performs data transmission in a single-link transmission manner, or may also adopt a multi-link transmission manner for data transmission.
  • the edge server uses a single-link transmission mode for data transmission, the edge server needs to select a communication link through step S203; that is, in this case, it can be determined that the value of K is equal to 1.
  • the edge server uses a multi-link transmission method for data transmission, the edge server needs to select at least two communication links through step S203; that is, in this case, it can be determined that the value of K is greater than 1.
  • the specific implementation manner adopted by the edge server for step S203 is also different, please refer to the following description for details:
  • step S203 may be as follows: First, multiple communication links may be screened according to the network quality of each communication link according to the network quality optimization objective. Specifically, when the network quality optimization goal includes the packet loss rate to be achieved, and the network quality of the communication link includes the packet loss rate of the communication link, it can be selected from multiple communication links that the packet loss rate is less than or equal to the required packet loss rate. A communication link with a packet loss rate. When the network quality optimization goal includes the delay time to be achieved, and the network quality of the communication link includes the delay time of the communication link, the communication link whose delay time is less than or equal to the delay time to be achieved can be screened out from multiple communication links road.
  • the network quality optimization goal includes both the packet loss rate and the delay time to be achieved
  • the network quality of the communication link includes the packet loss rate of the communication link and the delay time of the communication link
  • multiple communication links can be selected. Filter out the communication links whose packet loss rate is less than or equal to the required packet loss rate and whose delay time is less than or equal to the required delay time.
  • the selected communication link After the communication link is screened based on any of the above three screening methods, if the number of the selected communication links is 1, the selected communication link can be directly selected as the final K communication link. If the number of communication links screened out is greater than 1, a communication link can be randomly selected from the multiple communication links screened out as the final K communication links; For the public network communication cost of the link, the communication link with the lowest public network communication cost is selected from the multiple communication links screened out as the final K communication links.
  • the public network communication cost of the communication link includes: public network export cost and public network entrance cost; The cost required to use the public network entrance at the rate, such as the cost of the public network egress and the cost of the public network entrance can both be yuan/Mbps.
  • the lowest public network communication cost means that the sum of the public network egress cost and the public network ingress cost is the smallest.
  • Any public network egress cost can be obtained by the edge server from the operator that provides the public network egress corresponding to any public network egress cost.
  • any public network ingress cost can be obtained by the edge server from the operator that provides the public network egress Obtained from the operator of the public network entrance corresponding to the cost of any public network entrance.
  • the edge server can support multiple transmission modes.
  • the specific implementation method of step S203 can be as follows: first, according to the network quality optimization target and the network quality of each communication link, from multiple communication links Select H communication links that are compatible with each transmission mode in the road, and H is an integer greater than 1; secondly, based on the public network communication costs of each communication link that each transmission mode is adapted to, calculate each The transmission cost of each transmission mode; then, according to the transmission cost of each transmission mode, the transmission mode whose transmission cost meets the cost condition can be selected from multiple transmission modes; H communication links adapted from the selected transmission mode , select K communication links, K ⁇ H.
  • the number of communication links adapted to each transmission mode may be the same or different, which is not limited.
  • the various transmission modes mentioned above may include but not limited to: multiple concurrent transmission modes, and multiple redundant transmission modes.
  • the so-called multi-channel concurrent transmission mode refers to a mode in which at least two communication links are used to transmit signaling data, each communication link transmits a data block of signaling data, and each data block is different from each other.
  • the so-called multi-channel redundant transmission mode refers to: the mode in which each communication link transmits signaling data; or, the multi-channel redundant transmission mode refers to: using at least two communication links to transmit signaling data, each communication link The link transmits one data block of signaling data, and there is a pattern of redundant data (ie, duplicate data) between at least two data blocks.
  • the signal when the value of K is equal to 1, that is, when one communication link is selected from multiple communication links through step S203, the signal can be directly transmitted through the selected communication link. command data.
  • the value of K is greater than 1, that is, when at least two communication links are selected from multiple communication links through step S203, the aforementioned multiple redundant transmission mode or multiple concurrent transmission mode can be used , to transmit signaling data through the selected at least two communication links.
  • multiple communication links may be operator links between the edge server and the target cloud server, or may be links between the edge server and other cloud servers (such as nearby access cloud server); then, the end point of each communication link in the K communication links selected from multiple communication links may be the target cloud server, or other cloud servers (such as nearby cloud server).
  • the signaling data is transmitted through the K communication links, so that the signaling data can be directly transmitted to the target cloud server, as shown in Figure 3a
  • the signaling data can be transmitted through K communication links, which can realize the transmission of signaling data to Other cloud servers (such as a nearby cloud server), in this case, other cloud servers (such as a nearby cloud server) need to forward the signaling data to the target cloud server, as shown in Figure 3b.
  • the edge server in the embodiment of this application can perform link planning between the edge server and the target remote server according to the data transmission strategy between the edge server and the target remote server after the signaling data needs to reach the target cloud server , and select K communication links that meet the network quality optimization goal from the planned multiple communication links, so as to transmit the signaling data through the K communication links.
  • the planned communication links can be different with different data transmission strategies, which can effectively improve the flexibility of communication links performance, thereby improving the flexibility of data transmission; and, since the K communication links meet the network quality optimization goal, it can be explained that the network quality of the K communication links is relatively good, then the K communication links When transmitting signaling data, the reliability of data transmission can be effectively improved, thereby improving the effect of data transmission.
  • FIG. 4 is a schematic flow chart of a cloud-edge collaborative data transmission method provided by another embodiment of the present application.
  • the cloud-edge collaborative data transmission method can be executed by the aforementioned edge server, and the edge server can communicate with multiple cloud servers.
  • the edge server adopts a multi-link transmission mode for data transmission as an example; that is, the edge server in the embodiment of the present application can support multiple transmission modes, and the multiple transmission modes can include at least : Multiple redundant transmission mode and multiple concurrent transmission mode.
  • the data transmission method for cloud-edge collaboration may include the following steps S401-S407:
  • S402. Perform link planning between the edge server and the target cloud server according to the data transmission strategy between the edge server and the target cloud server to obtain multiple communication links.
  • the data transmission strategy is mainly used to instruct the edge server to transmit data to the target cloud server through the public network;
  • the intranet transmits data to the target cloud server.
  • the method of determining the data transmission strategy may be as follows: first, the edge server may determine the communication requirements of signaling data according to the geographic location of the edge server and the geographic location of the target cloud server. Specifically, if the geographic location of the edge server and the geographic location of the target cloud server are located in the same region, it can be determined that the communication requirements of the signaling data are communication requirements of the same region; If the geographical location of the target cloud server is not located in the same region, it can be determined that the communication requirement of the signaling data is a cross-regional communication requirement.
  • the data transmission strategy can be determined as: the strategy of the edge server transmitting data to the target cloud server through the public network; Let the data communication requirements be cross-regional communication requirements, then the edge server can directly determine the data transmission strategy as follows: the edge server connects to the cloud server nearby, and the cloud server connected nearby transmits the data to the target cloud server through the intranet Strategy.
  • the data transmission strategy may be determined according to the network optimization scheme.
  • the network optimization scheme can be a scheme that gives priority to transmission cost, or a scheme that gives priority to transmission performance; the transmission performance mentioned here can be measured by the reliability of the transmission process, and reliability and transmission performance can be positively correlated , that is, the higher the reliability, the better the transmission performance.
  • the network optimization scheme is a scheme that gives priority to the transmission cost, since the transmission cost required for public network transmission is usually less than the cost required for intranet transmission, the data transmission strategy can be determined as follows: the edge server will The strategy for data transmission to the target cloud server.
  • the data transmission strategy can be determined as follows: the edge server connects to the cloud server nearby, and the nearest The strategy for the connected cloud server to transmit data to the target cloud server through the intranet.
  • the network optimization scheme mentioned above may be directly given by the user through interface operations, or may be determined according to experience values or service requirements when the user has a user demand for network quality optimization.
  • the way for the user to directly specify the network optimization scheme through interface operations may include but not limited to: the terminal device used by the user may provide the user with a demand setting interface, and provide a scheme input area in the demand setting interface, so that The user can input a given network optimization scheme in the scheme input area.
  • the terminal device used by the user may provide the user with a requirement setting interface, and provide a solution selection area in the requirement setting interface, and the solution selection area includes the option of giving priority to the transmission cost and the solution giving priority to the transmission performance option, so that the user can select a solution option in the solution selection area to provide a network optimization solution.
  • the way to determine the network optimization scheme based on experience values or business needs can be: set a scheme that gives priority to transmission cost or a scheme that gives priority to transmission performance according to experience values or business needs in advance, as the default scheme.
  • the terminal device used by the user can provide the user with a network quality optimization component in any user interface, and the network quality optimization component can be understood as a "one-key optimization" component.
  • the terminal device used by the user can send the scheme identifier of the preset default scheme to the edge server, and the edge server will use the scheme indicated by the scheme identifier as the network optimization scheme; or, if When the network quality optimization component is triggered, the terminal device used by the user can send a component trigger notification to the edge server, so that the edge server determines the preset default solution as the network optimization solution after receiving the component trigger notification.
  • H is an integer greater than 1.
  • the network quality optimization target can be directly given by the user through interface operations, or can be determined according to experience values or business requirements when the user has a user demand for network quality optimization.
  • the way for the user to directly set the network quality optimization goal through interface operations may include but not limited to: the terminal device used by the user can provide the user with a demand setting interface, and provide a delay time setting area in the demand setting interface And the packet loss rate setting area, so that the user can perform at least one of the following operations to achieve the setting of the network quality optimization goal: set the delay time to be achieved in the delay time setting area, and set the required loss rate in the packet loss rate equipment area. packet rate.
  • the way to determine the network quality optimization goal according to the experience value or business demand can be: at least one of the default packet loss rate and the default delay time can be set in advance according to the experience value or business demand. one item.
  • the terminal device used by the user can provide the user with a network quality optimization component in any user interface. If the network quality optimization component is triggered, the default packet loss rate can be used as the packet loss rate to be achieved, or the default delay time can be added to the network quality optimization target as the delay time to be achieved, or the default The packet rate is used as the packet loss rate to be achieved, and the default delay time is added to the network quality optimization goal as the delay time to be achieved.
  • the network quality optimization target mentioned in the embodiment of the present application may include at least one of the following: the delay time to be achieved, and the packet loss rate to be achieved.
  • step S403 may include: in any transmission mode, from multiple communication links Initially select H communication links; determine the minimum delay duration from the delay durations of each communication link in the initially selected H communication links; if the minimum delay duration is less than the required delay duration, then the The initially selected H communication links are determined to be H communication links suitable for any transmission mode; if the minimum delay time is not less than the required delay time, the edge server can re-execute the above steps until it is determined H communication links suitable for any transmission mode are generated.
  • the embodiment of the present application is only an example to illustrate the specific manner of selecting H communication links according to the delay time length, and is not exhaustive.
  • the initially selected H communication links need to satisfy the following Condition: the minimum delay duration among the delay durations of the initially selected H communication links is less than the required delay duration; or, the average delay duration of the initially selected H communication links is less than the required delay duration; Alternatively, the delay duration of each of the initially selected H communication links is less than the desired delay duration, and so on.
  • the following description will be made by taking the condition that the minimum delay duration among the delay durations of the initially selected H communication links is less than the required delay duration as an example.
  • step S403 can be Including: initially selecting H communication links from a plurality of communication links; performing mean value calculation on the packet loss rate of each communication link in the initially selected H communication links to obtain an average packet loss rate; s If the average packet loss rate is less than the desired packet loss rate, the initially selected H communication links are determined as H communication links that are compatible with the multi-channel concurrent transmission mode; if the average packet loss rate is not less than If the packet loss rate needs to be achieved, the edge server can re-execute the above steps until H communication links that are compatible with the multi-channel concurrent transmission mode are determined.
  • the initially selected H communication links when the initially selected H communication links are used as the H communication links compatible with the multi-channel concurrent transmission mode, the initially selected H communication links
  • the specific implementation manner of step S403 It may include: initially selecting H communication links from a plurality of communication links; using a comprehensive packet loss rate calculation formula, according to the packet loss rate of each of the initially selected H communication links, calculating The comprehensive packet loss rate of the initially selected H communication links; if the comprehensive packet loss rate of the initially selected H communication links is less than the required packet loss rate, then the initially selected H communication links, Determine the H communication links that are compatible with the multi-path redundant transmission mode; if the comprehensive packet loss rate is not less than the required packet loss rate, the edge server can re-execute the above steps until it is determined that it is compatible with the multi-path redundant transmission mode.
  • the initially selected H communication links that match the transmission mode. That is to say, in the multiple redundant transmission mode, when the initially selected H communication links are used as the H communication links compatible with the multiple redundant transmission mode, the initially selected H communication links
  • the link needs to meet the following conditions: the comprehensive packet loss rate calculated according to the packet loss rate of each communication link initially selected is less than the required packet loss rate.
  • L i represents the packet loss rate of the i-th communication link selected initially, i ⁇ [1, H]; ⁇ represents the operation of the continuous product.
  • the network quality optimization goal includes: the delay time to be achieved and the packet loss rate to be achieved
  • the network quality of the communication link includes the delay time of the communication link and the packet loss rate of the communication link
  • H communication links are initially selected When the conditions shown in the following formulas 1.1-1.3 are met, the H communication links initially selected can be used as H communication links compatible with the multi-channel concurrent transmission mode:
  • the edge server (indicated by the letter a) connects to the cloud server (indicated by the letter c) nearby, and the cloud server connected nearby transmits the data to the target cloud server (indicated by the letter c) through the intranet letter b)
  • the initially selected H communication links can be used as H communication links compatible with concurrent transmission mode:
  • L i_ab represents the packet loss rate of the i-th communication link among the initially selected H communication links between the edge server and the target cloud server; The packet loss rate of the i-th communication link among the initially selected H communication links between the cloud server and the nearest connected cloud server; L t represents the packet loss rate to be achieved, and D t represents the delay time to be achieved .
  • H communication links are initially selected When the road meets the conditions shown in the following formulas 2.1-2.3, the initially selected H communication links can be used as H communication links that are compatible with the multi-channel redundant transmission mode:
  • the edge server (indicated by the letter a) connects to the nearest cloud server (indicated by the letter c), and the nearby cloud server will transmit the data to the target cloud server (indicated by the letter b) through the intranet Indicates)
  • the initially selected H communication links can be used as the multi-channel communication link H communication links that match the redundant transmission mode:
  • (1-L i_ab ) represents between the edge server and the target cloud server, the success rate of the i-th communication link among the initially selected H communication links;
  • (1-L i_ab ) L i_ac ) represents the packet loss rate of the i-th communication link among the initially selected H communication links between the edge server and the nearby cloud server.
  • the edge server can calculate the successful transmission of each communication link adapted to any transmission mode according to the transmission rate of signaling data and the packet loss rate of each communication link adapted to any transmission mode The data transfer rate required respectively.
  • the difference between the reference value (such as the value 1) and the packet loss rate of each communication link adapted to any transmission mode can be calculated to obtain each communication link adapted to any transmission mode.
  • the transmission rate of signaling data can be divided by the success rate of each communication link adapted to any transmission mode to obtain the successful transmission of each communication link adapted to any transmission mode The data transfer rate required respectively.
  • the edge server After obtaining the transmission rate required for each communication link adapted to any transmission mode to successfully transmit data, the edge server can adapt the transmission rate required for each communication link according to any transmission mode, and The public network communication cost of each communication link adapted to any transmission mode, calculate the public network communication cost of any transmission mode.
  • the public network communication cost of the communication link includes: the public network egress cost and the public network ingress cost; then when the edge server calculates the public network communication cost of any transmission mode, it can The transmission rate required for each communication link adapted, and the public network egress cost of each communication link adapted to any transmission mode, calculate the total public network egress cost of any transmission mode; the specifics can be calculated The product of the transmission rate required for each communication link adapted to any transmission mode and the corresponding public network egress cost is obtained to obtain the transmission public network entry cost of each communication link adapted to any transmission mode , sum the transmission public network entrance costs of each communication link adapted to any transmission mode, and obtain the total public network exit cost of any transmission mode.
  • the total cost of any transmission mode can be calculated according to the required transmission rate of each communication link adapted to any transmission mode, and the public network entrance cost of each communication link adapted to any transmission mode.
  • Public network entrance cost specifically, the product of the transmission rate required for each communication link adapted to any transmission mode and the corresponding public network entrance cost can be calculated to obtain each communication adapted to any transmission mode
  • the transmission public network entrance cost of the link is summed to obtain the total public network entrance cost of any transmission mode.
  • the total public network egress cost and the total public network entrance cost can be summed to obtain the public network communication cost of any transmission mode.
  • the edge server can calculate the transmission cost of any transmission mode based on the public network communication cost of any transmission mode. Specifically, if the data transmission strategy is used to indicate: the edge server transmits data to the target cloud server through the public network; then based on the public network communication cost of any transmission mode, the specific implementation of the step of determining the transmission cost of any transmission mode The method may be: determining the public network communication cost of any transmission mode as the transmission cost of any transmission mode.
  • the edge server connects to the cloud server nearby, and the cloud server connected nearby transmits the data to the target cloud server through the intranet; then based on the public network communication cost of any transmission mode, determine any
  • the specific implementation of the step of the transmission cost of the transmission mode can be: according to the intranet communication cost and transmission rate under the unit rate, calculate the total intranet communication cost; specifically, the intranet communication cost under the unit rate and the The product of the transmission rate is used as the total intranet communication cost. Then, the total intranet communication cost and the public network communication cost of any transmission mode can be summed to obtain the transmission cost of any transmission mode.
  • any of the above-mentioned transmission modes may be a multiple redundant transmission mode or a multiple concurrent transmission mode.
  • the transmission cost of any transmission mode can be shown in the following formula 3.1; when the data transmission strategy is used to indicate: the edge server is closest to When the cloud server is connected to the cloud server and the nearby cloud server transmits the data to the target cloud server through the intranet, the transmission cost of any transmission mode can be shown in the following formula 3.2.
  • mincost represents the transmission cost
  • S represents the transmission rate of the signaling data (in Mbps)
  • P i_ab represents the i-th rate that any transmission mode is adapted to between the edge server and the target cloud server.
  • the public network egress cost of each communication link (unit/Mbps)
  • P i_ac represents the public network egress of the i-th communication link adapted to any transmission mode between the edge server and the nearby cloud server Cost (unit unit/Mbps)
  • Q i_ab indicates the public network entrance cost (unit unit/Mbps) of the i-th communication link to which any transmission cost is adapted
  • B indicates the intranet communication cost (unit unit/Mbps).
  • each transmission mode selects a transmission mode whose transmission cost satisfies the cost condition from multiple transmission modes.
  • the cost condition can be set according to business requirements or experience values.
  • the cost condition may be the minimum transmission cost; then, when the edge server executes step S405, it can select the transmission mode with the minimum transmission cost from multiple transmission modes according to the transmission cost of each transmission mode.
  • the cost condition may be a condition that the transmission cost is less than the cost threshold; then, when the edge server executes step S405, it can select any transmission mode with a transmission cost less than the cost threshold from multiple transmission modes according to the transmission cost of each transmission mode. - transfer mode.
  • K is an integer greater than 1, and K ⁇ H.
  • the edge server may directly determine the H communication links adapted to the selected transmission mode as K communication links. In another implementation manner, the edge server may randomly select K communication links from the H communication links adapted to the selected transmission mode. In another embodiment, the edge server can follow the relationship that the packet loss rate (or delay time) is proportional to the selection order, according to the packet loss rate (or delay time) of each communication link adapted to the selected transmission mode ), from the H communication links adapted to the selected transmission mode, select K communication links; the so-called packet loss rate (or delay) is proportional to the relationship between selection order means: the communication link The smaller the packet loss rate (or the delay time), the more preferentially the communication link is selected.
  • the edge server may also comprehensively consider the delay time and packet loss rate of each communication link adapted to the selected transmission mode, and select K from the H communication links adapted to the selected transmission mode Communication link; in this case, each of the selected K communication links needs to satisfy: the packet loss rate is less than the packet loss rate threshold, and the delay time is less than the time length threshold. It should be understood that the embodiment of the present application only lists several ways of selecting K communication links as examples, and is not exhaustive.
  • the edge server may adopt the selected transmission mode and allocate a data packet to each of the K communication links according to the signaling data. Specifically, if the selected transmission mode is the multiplex redundant transmission mode, the edge server may assign the signaling data as a data packet to each of the K communication links respectively; that is, In this case, the data packets allocated to each communication link are essentially signaling data. If the selected transmission mode is multi-channel concurrent transmission mode, the edge server can block the signaling data to obtain K data packets; distribute K data packets to K communication links, and one communication link is One data packet is allocated; that is, the data packet allocated to each communication link in this case is essentially a data block of signaling data, and the data packets allocated to each communication link are different from each other.
  • the edge server can respectively transmit corresponding data packets through each of the K communication links.
  • the edge server may transmit corresponding data packets directly through each of the K communication links.
  • the edge server can determine the mode mark of the selected transmission mode, and add the mode mark to the data packet corresponding to each of the K communication links; specifically, the data packet can be Add this schema tag to the header in the .
  • the corresponding data packets added with the mode mark are transmitted respectively; by adding the mode mark, it is convenient to inform the receiving end of the data packet how to process the received data packet, so as to Get signaling data.
  • the added mode mark can be represented by R; when the selected transmission mode is the multiple redundant transmission mode, the added mode mark can be represented by M express.
  • the edge server in the embodiment of the present application can determine the data transmission strategy between the edge server and the target remote server according to the network optimization scheme that meets the user's actual communication needs after the signaling data needs to reach the target cloud server.
  • the data transmission strategy that meets the user's needs combined with the multiple operator links established between the edge server and the cloud server, and the reliable communication link of the cloud intranet, the link between the edge server and the target remote server Road planning, so that the planned multiple communication links meet user needs.
  • the planned multiple communication links it can also provide multiple transmission modes according to the network quality optimization target that meets the needs of users. By solving the transmission costs of various transmission modes, it is possible to select from multiple communication links that meet the requirements of the network.
  • the K communication links with the network quality optimization target and the lowest transmission cost are used to transmit the signaling data.
  • Such a transmission method can not only optimize the transmission method according to user needs, but also realize the data transmission effect to meet user needs and minimize transmission costs, thereby improving user experience and user stickiness;
  • the combination of the transmission method and the reliable network transmission capability of the cloud intranet greatly improves the reliability of cloud-edge collaboration and reduces the cost of collaboration.
  • this embodiment of the present application also proposes a cloud-side collaborative data transmission method.
  • the cloud-edge collaborative data transmission method may be executed by a current cloud server, and the current cloud server refers to a cloud server that communicates with an edge server through K communication links.
  • the data transmission method for cloud-edge collaboration may include the following steps S501-S503:
  • S501 Receive signaling data transmitted by an edge server through K communication links.
  • K is a positive integer.
  • the K communication links are selected from multiple communication links according to the network quality optimization goal; the multiple communication links are selected between the edge server and the target cloud server according to the data transmission strategy between the edge server and the target cloud server. It is obtained by performing link planning between target cloud servers, where the target cloud server refers to the cloud server to which the signaling data needs to reach.
  • the edge server if the value of K is equal to 1, it indicates that the edge server directly transmits signaling data through one communication link; then, when the current cloud server in this case executes step S501, it can directly receive the edge The signaling data transmitted by the server through this 1 communication link. If the value of K is greater than 1, it indicates that the edge server transmits K data packets related to signaling data through at least two communication links; then, when the current cloud server in this case executes step S501, it can receive The edge server transmits K data packets through K communication links, and one communication link transmits one data packet.
  • the current cloud server receiving signaling data through the K communication links may be the target cloud server, or a cloud server connected to the edge server nearby; based on different situations, the current cloud server requires The data processing operations performed are different. Based on this, the current cloud server can determine whether the destination address of the signaling data points to the current cloud server by detecting whether the destination address of the signaling data is the public network communication address of the current cloud server. If the destination address of the signaling data is the public network communication address of the current cloud server, it means that the destination address of the signaling data points to the current cloud server, that is, the current cloud server is the target cloud server, and step S502 can be executed at this time.
  • step S503 can be executed at this time.
  • step S502 may be: if the destination address of the received signaling data points to the current cloud server, then store the signaling data, Or perform response processing on the signaling data.
  • the response processing to the signaling data may be: determine the game screen to be displayed according to the user operation data, and send the game screen to the game client for further processing. show.
  • step S502 may be: if the destination address of the received signaling data points to the current cloud server, then integrate the K data packets to obtain the signaling data. Specifically, the current cloud server can parse out the mode mark from the K data packets. If the parsed mode mark is the mode mark of the multi-channel redundant transmission mode, the redundant data in the K data packets are fused to obtain the signaling data; optionally, if the K data packets have bit errors, the data packets can also be Based on data packets other than the errored data packet among the K data packets, the errored data packet is recovered to obtain signaling data.
  • the K data packets are rearranged to obtain the signaling data; specifically, each data packet can have a reordering number, which can be reordered according to The K data packets are rearranged in ascending order of numbers to obtain signaling data.
  • step S503 may be: if the received signaling data does not point to the current cloud server, then determine the destination address based on the signaling data The intranet communication address of the target cloud server directly forwards the signaling data to the target cloud server through the determined intranet communication address.
  • step S503 can be: if the received signaling data does not point to the current cloud server, then K data packets are forwarded to the target cloud server through the intranet Specifically, the intranet communication address of the target cloud server may be determined based on the destination address of the signaling data, and the K data packets may be forwarded to the target cloud server through the determined intranet communication address.
  • the multiple communication links in the embodiment of the present application are obtained through dynamic link planning based on data transmission strategies. Therefore, with different data transmission strategies, the planned multiple communication links can be different, which can effectively improve The flexibility of communication links, thereby improving the flexibility of data transmission; and, since the K communication links meet the network quality optimization goal, it can be explained that the network quality of these K communication links is relatively good, then through this When K communication links transmit signaling data, the reliability of data transmission can be effectively improved, thereby improving the effect of data transmission.
  • the cloud-edge collaborative data transmission method proposed in the embodiment of this application can be applied to various scenarios, such as cloud game scenarios, common game scenarios, audio and video playback scenarios, browser-based information browsing scenarios, etc. Wait.
  • cloud game scenarios such as cloud game scenarios, common game scenarios, audio and video playback scenarios, browser-based information browsing scenarios, etc. Wait.
  • video rendering servers that is, servers used to render game screens
  • the game server the cloud server used to provide game services
  • the edge server can adopt the data transmission scheme of cloud-side collaboration, and comprehensively consider multi-chain for different communication needs. Based on factors such as communication quality and cost of the road, an optimized transmission mode is planned for signaling data for data transmission and user experience is improved.
  • the general process of using the cloud-side collaborative data transmission method in the cloud game scene is as follows:
  • the game client can output a demand setting interface 60 for the user, and the demand setting interface 60 can support the user to set information such as network optimization schemes and network quality optimization goals.
  • the requirement setting interface 60 may include a scheme selection area 61 and a network quality input area 62, as shown in FIG.
  • the network quality input area 62 may include a packet loss rate setting area 621 and a delay time setting area 622.
  • the user can select a target program option in the program selection area 61, and set the packet loss rate to be achieved in the packet loss rate setting area 621 respectively, and set the delay time to be reached in the delay time setting area 622; Then, the user can trigger the demand confirmation component 63 in the demand setting interface, so that the game client will determine the scheme indicated by the target scheme option selected by the user as the cost optimization scheme given by the user, and adopt the desired cost optimization scheme set by the user.
  • the packet loss rate and the delay time to be achieved generate a network quality optimization goal given by the user, and upload the cost optimization plan and network quality optimization goal to the edge server, as shown in Figure 6a.
  • the game client can generate signaling data according to the user's game operation behavior, and send the signaling data to the edge server.
  • the edge server receives the signaling data sent by the game client used by the user, it can first determine the target cloud server (that is, the target game server) that the signaling data to be transmitted needs to reach, and according to the location of the edge server The geographic location of the target cloud server and the geographic location of the target cloud server determine the communication requirements of the signaling data. Then, based on the different communication requirements of signaling data, different orchestration logics can be used to arrange the transmission mode, as follows:
  • the edge server can transmit multiple operator links between the edge server and the target cloud server according to the strategy of the edge server transmitting data to the target cloud server through the public network As multiple communication links, then based on the public network communication cost and network quality of each communication link, K communication links are comprehensively selected, and multiple redundant transmission modes or multiple concurrent transmission modes are adopted, according to the signaling data Assign a data packet to each of the K communication links, and then transmit their respective data packets based on each of the K communication links, so as to transmit the signaling data to the target cloud server.
  • the edge server may further determine the cost optimization solution selected by the user. If the cost optimization scheme selected by the user is a scheme that gives priority to transmission costs, the edge server can preferentially select multiple links with the highest network quality among the multi-links for public network forwarding, that is, the edge server can choose the edge server to pass through the public network The strategy of transmitting data to the target cloud server; if the cost optimization solution selected by the user is a solution that prioritizes transmission performance, the edge server can give priority to accessing the nearest cloud server and forward it through the intranet via the nearby cloud server , that is, the edge server can choose the strategy that the edge server connects to the cloud server nearby, and the nearby cloud server transmits the data to the target cloud server through the intranet.
  • the edge server can carry out link planning according to the data transmission strategy, and based on the public network communication cost and network quality of each communication link, perform multiple selection to select K communication links, and then , using the multiple redundant transmission mode or the multiple concurrent transmission mode, according to the signaling data, a data packet is assigned to each of the K communication links, and then each communication link based on the K communication links Each path transmits its own data packets, so as to realize the transmission of the signaling data to the target cloud server.
  • the current cloud server that communicates with the edge server through K communication links can use the operation flowchart shown in Figure 6b to perform related operations:
  • step b Perform different operations according to the mode flag in the packet header of the current data packet. If the mode mark in the packet header is R (i.e. the mode mark of the multiplex redundant transmission mode), then jump to step a); if the mode mark in the packet header is M (i.e. the mode mark of the multiplex concurrent transmission mode), then Jump to step b);
  • a) Summarize the data packets transmitted by the K communication links, and try to fuse the redundant data in the K data packets, and recover the erroneous data packets in the K data packets, so as to recover the signaling data. If the signaling data is successfully restored, feed back an ACK (Acknowledgment, confirmation) packet to the edge server, and jump to step 4); if the signaling data is not successfully restored, feed back a NACK (Negative Acknowledgment, denial of confirmation) to the edge server packet, and jump to step 1);
  • the cloud-edge collaborative data transmission scheme proposed in the embodiment of the present application in the cloud game scene, it can be realized how to stably and reliably transmit the user's signaling data to the cloud game server in the cloud game scene, so that the game server The rendered video stream can be sent back in time, thereby improving the user's gaming experience.
  • it can also realize the function of optimizing the transmission method according to user needs, improve the reliability of game control command transmission for cloud game users, reduce control delay, and improve user game experience.
  • the reliability of cloud-side collaboration in cloud games can be greatly improved and the cost of collaboration can be reduced.
  • the embodiment of the present application also discloses a cloud-edge collaborative data processing device, and the cloud-edge collaborative data processing device may be A computer-readable instruction (including program code) running on an edge server.
  • the data processing device for cloud-edge collaboration may execute the method shown in FIG. 2 or FIG. 4 .
  • the data processing device for cloud-side collaboration can run the following units:
  • a determining unit 701 configured to determine a target cloud server where the signaling data to be transmitted needs to arrive;
  • the processing unit 702 is configured to perform link planning between the edge server and the target cloud server according to the data transmission strategy between the edge server and the target cloud server, to obtain multiple communication links;
  • the processing unit 702 is further configured to select K communication links satisfying the network quality optimization goal from the plurality of communication links according to the network quality optimization goal, where K is a positive integer;
  • a transmission unit 703, configured to transmit the signaling data through the K communication links.
  • the data transmission policy is used to indicate: the edge server transmits data to the target cloud server through the public network; Between the data transmission strategy, link planning is performed between the edge server and the target cloud server, and when multiple communication links are obtained, it can be specifically used for:
  • the multiple operator links between the edge server and the target cloud server are used as multiple communication links.
  • the data transmission policy is used to indicate that: the edge server accesses the cloud server nearby, and the cloud server connected nearby transmits data to the target cloud server through the intranet; correspondingly,
  • the processing unit 702 is configured to perform link planning between the edge server and the target cloud server according to the data transmission strategy between the edge server and the target cloud server to obtain multiple communication links, it may specifically Used for:
  • the plurality of operator links between the edge server and the cloud server connected to the edge server nearby are used as the plurality of communication links.
  • the edge server is configured with n public network communication addresses
  • the cloud server is configured with m public network communication addresses, and both n and m are positive integers;
  • An operator link is established by using a public network communication address of the edge server and a public network communication address of the cloud server.
  • the edge server supports multiple transmission modes; correspondingly, the processing unit 702 is configured to select K communication links that meet the network quality optimization target from the multiple communication links according to the network quality optimization target.
  • the processing unit 702 is configured to select K communication links that meet the network quality optimization target from the multiple communication links according to the network quality optimization target.
  • For communication links it can be used specifically for:
  • the transmission cost of each transmission mode select a transmission mode whose transmission cost satisfies a cost condition from the multiple transmission modes;
  • the cost condition includes: a condition that the transmission cost is the smallest, or a condition that the transmission cost is less than a cost threshold;
  • the network quality optimization target includes: a packet loss rate to be achieved, and the network quality of the communication link includes the packet loss rate of the communication link;
  • the multiple transmission modes include multiple concurrent transmission modes, and the processing unit 702 is used to select a transmission mode corresponding to each transmission mode from the multiple communication links according to the network quality optimization target and the network quality of each communication link.
  • Adapted H communication links can be specifically used for:
  • the average packet loss rate is less than the required packet loss rate, then determine the initially selected H communication links as H communication links that are compatible with the multi-channel concurrent transmission mode .
  • the network quality optimization target includes: a packet loss rate to be achieved, and the network quality of the communication link includes the packet loss rate of the communication link;
  • the multiple transmission modes include a multiple redundant transmission mode, and the processing unit 702 is used to select from the multiple communication links according to the network quality optimization target and the network quality of each communication link.
  • the processing unit 702 is used to select from the multiple communication links according to the network quality optimization target and the network quality of each communication link.
  • the comprehensive packet loss rate of the initially selected H communication links is less than the required packet loss rate, then the initially selected H communication links are determined to be compatible with the multi-path redundancy H communication links matching the transmission mode;
  • L i represents the packet loss rate of the i-th communication link selected initially, i ⁇ [1, H].
  • the network quality optimization target includes: the delay time to be achieved, and the network quality of the communication link includes the delay time of the communication link; the processing unit 702 is used to optimize the target according to the network quality and each communication
  • the network quality of the link when selecting H communication links suitable for each transmission mode from the plurality of communication links, can be specifically used for:
  • H communication links are initially selected from the plurality of communication links
  • the initially selected H communication links are determined as the H communication links suitable for any of the transmission modes.
  • processing unit 702 when used to calculate the transmission cost of each transmission mode based on the communication cost of each communication link adapted to each transmission mode, it may be specifically used for:
  • the transmission cost of any transmission mode is calculated.
  • the public network communication cost of the communication link includes: public network egress cost and public network ingress cost; the public network egress cost refers to the cost required to use the public network egress at a unit rate, and the The unit price of the public network entrance cost refers to the cost required to use the public network entrance at a unit rate;
  • the processing unit 702 calculates the required transmission rate for each communication link adapted according to any of the transmission modes, and the public network of each communication link adapted for any of the transmission modes.
  • Communication cost when calculating the public network communication cost of any of the transmission modes, can be specifically used for:
  • a summing operation is performed on the total public network egress cost and the total public network ingress cost to obtain the public network communication cost of any transmission mode.
  • the data transmission policy is used to indicate: the policy for the edge server to transmit data to the target cloud server through the public network;
  • the processing unit 702 when used to determine the transmission cost of any transmission mode based on the public network communication cost of any transmission mode, it may be specifically configured to: transmit the public network communication cost of any transmission mode cost, determined as the transmission cost of any one of the transmission modes.
  • the data transmission policy is used to indicate that: the edge server connects to a nearby cloud server, and the nearby connected cloud server transmits data to the target cloud server through an intranet;
  • the processing unit 702 when used to determine the transmission cost of any transmission mode based on the public network communication cost of any transmission mode, it may be specifically used for:
  • a sum operation is performed on the total intranet communication cost and the public network communication cost of any transmission mode to obtain the transmission cost of any transmission mode.
  • the transmission unit 703 when used to transmit the signaling data through the K communication links, it may be specifically used for:
  • Corresponding data packets are respectively transmitted through each of the K communication links.
  • the selected transmission mode is a multiplex redundant transmission mode; correspondingly, the transmission unit 703 is used to adopt the selected transmission mode, and according to the signaling data is the K
  • the transmission unit 703 is used to adopt the selected transmission mode, and according to the signaling data is the K
  • each communication link in the communication links allocates a data packet, it can be specifically used to: use the signaling data as a data packet and distribute it to each communication link in the K communication links respectively .
  • the selected transmission mode is a multi-channel concurrent transmission mode; correspondingly, the transmission unit 703 is used to adopt the selected transmission mode, according to the signaling data, for the K When each of the communication links allocates a data packet, it can be used specifically for:
  • the transmission unit 703 when used to respectively transmit corresponding data packets through each of the K communication links, it may be specifically used for:
  • the corresponding data packets added with the mode mark are respectively transmitted through each of the K communication links.
  • processing unit 702 may also be used to:
  • determining the data transmission strategy is: the strategy of the edge server transmitting data to the target cloud server through the public network;
  • the data transmission strategy is determined according to a network optimization scheme.
  • processing unit 702 when used to determine the data transmission strategy according to the network optimization scheme, it may be specifically used to:
  • determining the data transmission strategy is: the strategy for the edge server to transmit data to the target cloud server through the public network;
  • the data transmission strategy is determined as follows: the edge server connects to the cloud server nearby, and the cloud server connected nearby transmits the data to the target cloud through the intranet Server policy.
  • each unit in the cloud-edge collaborative data transmission device shown in FIG. can also be divided into multiple functionally smaller units, which can achieve the same operation without affecting the realization of the technical effects of the embodiments of the present application.
  • the above-mentioned units are divided based on logical functions.
  • the functions of one unit may also be realized by multiple units, or the functions of multiple units may be realized by one unit.
  • the data transmission device based on cloud-side collaboration may also include other units.
  • these functions may also be implemented with the assistance of other units, and may be implemented by cooperation of multiple units.
  • a general-purpose computing device such as a computer including processing elements such as a central processing unit (CPU), a random access storage medium (RAM), and a read-only storage medium (ROM) and storage elements.
  • Executing computer-readable instructions capable of executing the steps involved in the corresponding method as shown in Figure 2 or Figure 4, to construct a cloud-edge collaborative data transmission device as shown in Figure 7, and Realize the game processing method of the embodiment of the present application.
  • the computer-readable instructions may be recorded in, for example, a computer-readable recording medium, loaded into the above-mentioned computing device via the computer-readable recording medium, and executed therein.
  • the edge server in the embodiment of this application can perform link planning between the edge server and the target remote server according to the data transmission strategy between the edge server and the target remote server after the signaling data needs to reach the target cloud server , and select K communication links that meet the network quality optimization goal from the planned multiple communication links, so as to transmit the signaling data through the K communication links.
  • the planned communication links can be different with different data transmission strategies, which can effectively improve the flexibility of communication links performance, thereby improving the flexibility of data transmission; and, since the K communication links meet the network quality optimization goal, it can be explained that the network quality of the K communication links is relatively good, then the K communication links When transmitting signaling data, the reliability of data transmission can be effectively improved, thereby improving the effect of data transmission.
  • the embodiment of the present application also discloses a cloud-edge collaborative data processing device.
  • the data processing device for cloud-edge collaboration can execute the method shown in FIG. 5 .
  • the data processing device for cloud-edge collaboration can run the following units:
  • the transmission unit 801 is configured to receive signaling data transmitted by the edge server through K communication links, K is a positive integer; the K communication links are selected from multiple communication links according to the network quality optimization goal; The multiple communication links are obtained by performing link planning between the edge server and the target cloud server according to the data transmission strategy between the edge server and the target cloud server, and the target cloud server is Refers to the cloud server to which the signaling data needs to arrive;
  • the processing unit 802 is configured to perform data processing on the received signaling data if the destination address of the received signaling data points to the current cloud server, and the current cloud server refers to: through the K communication links A cloud server communicating with the edge server;
  • the transmission unit 801 is further configured to forward the received signaling data to the target cloud server through an intranet if the received signaling data is not directed to the current cloud server.
  • the transmission unit 801 when used to receive signaling data transmitted by the edge server through K communication links, it may be specifically configured to: receive K data packets transmitted by the edge server through K communication links;
  • the processing unit 802 When the processing unit 802 is used for performing data processing on the received signaling data if the destination address of the received signaling data points to the current cloud server, it can be specifically used for: if the destination of the received signaling data The address points to the current cloud server, then the K data packets are integrated to obtain the signaling data;
  • the transmission unit 801 When the transmission unit 801 is used to forward the received signaling data to the target cloud server through the intranet if the received signaling data does not point to the current cloud server, it may specifically use In: if the received signaling data does not point to the current cloud server, then forward the K data packets to the target cloud server through the intranet.
  • processing unit 802 when used to integrate the K data packets to obtain the signaling data, it may be specifically used for:
  • the parsed mode mark is the mode mark of the multiplex redundant transmission mode, then fusing the redundant data in the K data packets to obtain the signaling data;
  • the K data packets are rearranged to obtain the signaling data.
  • each unit in the cloud-edge collaborative data transmission device shown in FIG. can also be divided into multiple functionally smaller units, which can achieve the same operation without affecting the realization of the technical effects of the embodiments of the present application.
  • the above-mentioned units are divided based on logical functions.
  • the functions of one unit may also be realized by multiple units, or the functions of multiple units may be realized by one unit.
  • the data transmission device based on cloud-side collaboration may also include other units.
  • these functions may also be implemented with the assistance of other units, and may be implemented by cooperation of multiple units.
  • a general-purpose computing device such as a computer including processing elements such as a central processing unit (CPU), a random access storage medium (RAM), and a read-only storage medium (ROM) and storage elements.
  • CPU central processing unit
  • RAM random access storage medium
  • ROM read-only storage medium
  • Run computer-readable instructions capable of executing the steps involved in the corresponding method as shown in Figure 5, to construct a cloud-edge collaborative data transmission device as shown in Figure 8, and to realize the present invention Apply for the game processing method of the embodiment.
  • the computer-readable instructions may be recorded in, for example, a computer-readable recording medium, loaded into the above-mentioned computing device via the computer-readable recording medium, and executed therein.
  • the multiple communication links in the embodiment of the present application are obtained through dynamic link planning based on data transmission strategies. Therefore, with different data transmission strategies, the planned multiple communication links can be different, which can effectively improve The flexibility of the communication link, thereby improving the flexibility of data transmission; and, since the K communication links meet the network quality optimization target given by the user, then when transmitting signaling data through these K communication links, the To make the data transmission effect meet the needs of users, the transmission of signaling data can be realized while meeting the needs of users, thereby improving user experience and user stickiness.
  • the embodiment of the present application further provides a server; the server may be the aforementioned edge server, or the aforementioned current cloud server.
  • the server at least includes a processor, an input interface, an output interface, and a computer storage medium; and the processor, input interface, output interface, and computer storage medium in the server can be connected through a bus or other methods.
  • the computer storage medium can be stored in the memory of the server, and the computer storage medium is used to store computer-readable instructions.
  • the processor or CPU (Central Processing Unit, central processing unit)
  • CPU Central Processing Unit, central processing unit
  • the processor described in the embodiment of the present application may be used to perform a series of cloud-edge collaborative data transmission methods.
  • the processor described in the embodiment of the present application can be used to perform a series of cloud-edge collaborative data transmission methods.
  • the embodiment of the present application also provides a computer storage medium (Memory).
  • the computer storage medium is a memory device in a server and is used to store programs and data. It can be understood that the computer storage medium here may include a built-in storage medium in the server, and certainly may include an extended storage medium supported by the server.
  • the computer storage medium provides storage space, and the storage space stores the operating system of the server. Moreover, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions may be one or more computer-readable instructions (including program codes).
  • the computer storage medium here can be a high-speed RAM memory, or a non-volatile memory (non-volatile memory), such as at least one disk memory; computer storage media.
  • the server shown in FIG. 9 is an edge server
  • one or more instructions stored in the computer storage medium may be loaded and executed by the processor to realize the above-mentioned cloud edge shown in FIG. 2 or 4. Corresponding steps of the method in the coordinated data transmission method embodiment.
  • the processor can load and execute one or more instructions stored in the computer storage medium, so as to realize the above-mentioned cloud-edge collaboration shown in FIG. 5 Corresponding steps of the method in the data transmission method embodiment.
  • a computer program product or computer program is also provided, where the computer program product or computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium.
  • the processor of the computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the embodiment of the cloud-edge collaborative data transmission method shown in FIG. 2 , FIG. 4 or FIG. 5 above.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

一种云边协同的数据传输方法,包括:确定待传输的信令数据需到达的目标云端服务器(S201);按照边缘服务器和目标云端服务器之间的数据传输策略,在边缘服务器和所述目标云端服务器之间进行链路规划,得到多条通信链路(S202);按照网络质量优化目标,从多条通信链路中选择满足网络质量优化目标的K条通信链路,K为正整数(S203);通过该K条通信链路传输信令数据(S204)。

Description

云边协同的数据传输方法、服务器及存储介质
本申请要求于2021年07月13日提交中国专利局,申请号为202110793799.4,申请名称为“云边协同的数据传输方法、装置、服务器及存储介质”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本申请涉及互联网技术领域,具体涉及云技术领域,尤其涉及一种云边协同的数据传输方法、装置、服务器及存储介质。
背景技术
云边协同是指边缘计算和云计算之间的紧密协同,边缘计算由边缘服务器执行,云计算则由云端服务器执行;在云边协同的过程中,边缘服务器可凭借“边缘”的特性,接收用户设备发送的信令数据,并将接收到的信令数据转发给相应的云端服务器,使得该云端服务器可根据该信令数据执行后续处理。目前,如何在边缘服务器和云端服务器之间传输信令数据成为了研究热点。
发明内容
本申请实施例提供了一种云边协同的数据传输方法,所述数据传输方法包括:
确定待传输的信令数据需到达的目标云端服务器;
按照边缘服务器和所述目标云端服务器之间的数据传输策略,在所述边缘服务器和所述目标云端服务器之间进行链路规划,得到多条通信链路;
按照网络质量优化目标,从所述多条通信链路中选择满足所述网络质量优化目标的K条通信链路,K为正整数;
通过所述K条通信链路传输所述信令数据。
本申请实施例提供了一种云边协同的数据传输方法,所述数据传输方法包括:
接收边缘服务器通过K条通信链路传输的信令数据,K为正整数;所述K条通信链路是按照网络质量优化目标,从多条通信链路中选择的;所述多条通信链路是按照所述边缘服务器和目标云端服务器之间的数据传输策略,在所述边缘服务器和所述目标云端服务器之间进行链路规划得到的,所述目标云端服务器是指所述信令数据需到达的云端服务器;
若接收到的信令数据的目的地址指向当前云端服务器,则对所述接收到的信令数据进行数据处理,所述当前云端服务器是指:通过所述K条通信链路和所述边缘服务器进行通信的云端服务器;
若所述接收到的信令数据未指向所述当前云端服务器,则通过内网将所述接收到的信令数据转发至所述目标云端服务器。
本申请实施例提供了一种云边协同的数据传输装置,所述数据传输装置包括:
确定单元,用于确定待传输的信令数据需到达的目标云端服务器;
处理单元,用于按照边缘服务器和所述目标云端服务器之间的数据传输策略,在所述边缘服务器和所述目标云端服务器之间进行链路规划,得到多条通信链路;
所述处理单元,还用于按照网络质量优化目标,从所述多条通信链路中选择满足所述网络质量优化目标的K条通信链路,K为正整数;
传输单元,用于通过所述K条通信链路传输所述信令数据。
本申请实施例提供了一种云边协同的数据传输装置,所述数据传输装置包括:
传输单元,用于接收边缘服务器通过K条通信链路传输的信令数据,K为正整数;所述K条通信 链路是按照网络质量优化目标,从多条通信链路中选择的;所述多条通信链路是按照所述边缘服务器和目标云端服务器之间的数据传输策略,在所述边缘服务器和所述目标云端服务器之间进行链路规划得到的,所述目标云端服务器是指所述信令数据需到达的云端服务器;
处理单元,用于若接收到的信令数据的目的地址指向当前云端服务器,则对所述接收到的信令数据进行数据处理,所述当前云端服务器是指:通过所述K条通信链路和所述边缘服务器进行通信的云端服务器;
所述传输单元,还用于若所述接收到的信令数据未指向所述当前云端服务器,则通过内网将所述接收到的信令数据转发至所述目标云端服务器。
本申请实施例提供了一种服务器,所述服务器包括输入接口和输出接口,所述服务器还包括:处理器,适于实现一条或多条计算机可读指令;以及,计算机存储介质,所述计算机存储介质存储有一条或多条计算机可读指令;所述一条或多条指令适于由所述处理器加载并执行上述的云边协同的数据传输方法。
本申请实施例提供了一种非易失性的计算机可读存储介质,所述非易失性的计算机可读存储介质存储有一条或多条计算机可读指令,所述一条或多条计算机可读指令适于由处理器加载并执行上述的云边协同的数据传输方法。
附图说明
为了更清楚地说明本申请实施例技术方案,下面将对实施例描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图是本申请的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1a是本申请实施例提供的一种云边协同的数据传输系统的系统架构图;
图1b是本申请实施例提供的一种云边协同的数据传输方案的方案示意图;
图2是本申请实施例提供的一种云边协同的数据传输方法的流程示意图;
图3a是本申请实施例提供的一种通过K条通信链路传输信令数据的传输示意图;
图3b是本申请实施例提供的另一种通过K条通信链路传输信令数据的传输示意图;
图4是本申请另一实施例提供的云边协同的数据传输方法的流程示意图;
图5是本申请另一实施例提供的云边协同的数据传输方法的流程示意图;
图6a是本申请实施例提供的一种发送成本优化方案和网络质量优化目标的示意图;
图6b是本申请实施例提供的一种云端服务器对接收到的数据包执行相关操作的流程示意图;
图7是本申请实施例提供的一种云边协同的数据传输装置的结构示意图;
图8是本申请另一实施例提供的一种云边协同的数据传输装置的结构示意图;
图9是本申请实施例提供的一种服务器的结构示意图。
具体实施方式
下面将结合本申请实施例中的附图,对本申请实施例中的技术方案进行清楚、完整地描述。
本申请实施例提出了一种基于多路选优的云边协同的数据传输方案,该数据传输方案可根据网络质量优化目标以及通信链路的网络质量,自适应地选择合适的数据传输策略(可简称为传输策略)和传输方式,并灵活地选择网络质量较优的通信链路进行数据传输,以实现在满足用户需求的同时,降低用户的成本,提升数据传输的灵活性,从而提升用户体验和用户粘度。
为便于实现该数据传输方案,本申请实施例提出了一种云边协同的数据传输系统;参见图1a所示,该数据传输系统可包括但不限于:部署在边缘的边缘服务器(或称为边缘节点),以及部署在不同地域 的云端服务器(或称为云节点)。其中,云端服务器之间可通过内网(或称为私网、局域网等)进行通信;边缘服务器可在公网(或称为外网、广域网、Internet(因特网)等)中,通过不同的运营商和每个云端服务器建立一条或多条运营商链路,并基于建立的运营商链路和相应的云端服务器进行通信。进一步的,边缘服务器可包括链路质量监控层、多路选优层以及传输层;该链路质量监控层主要负责搜集边缘服务器和各个云端服务器之间的各条运营商链路的网络质量,以便于后续编排分析数据传输策略和传输方式;多路选优层则主要负责编排分析出合适的数据传输策略,并根据网络质量优化目标,综合考虑链路质量监控层搜集到的各条运营商链路的网络质量以及各条运营商链路的通信成本,选择合适的传输方式以及且较优的运营商链路;而传输层则主要负责根据多路选优层所编排分析出的数据传输策略、传输方式以及多路选优层所选择的运营商链路进行数据传输。
需要说明的是,图1a只是示例性地表征数据传输系统的系统架构,并不对此进行限定。例如,图1a只是示例性地表征了一个边缘服务器,但在实际应用中,该数据传输系统所包括的边缘服务器的数量不局限于1个,也可以是2个、3个甚至更多;又如,图1a只是示例性地表征了位于地域1和地域2的两个云端服务器,但在实际应用中,该数据传输系统中还可包括更多数量的云端服务器;再如,图1a中的边缘服务器和其中一个云端服务器均位于地域1,但实际应用中,边缘服务器和各个云端服务器可位于不同的地域,等等。还需要说明的是,上述所提及的任一服务器(如边缘服务器、云端服务器)可以是是独立的物理服务器,也可以是多个物理服务器构成的服务器集群或者分布式系统,还可以是提供云服务、云数据库、云计算、云函数、云存储、网络服务、云通信、中间件服务、域名服务、安全服务、CDN(Content Delivery Network,内容分发网络)、以及大数据和人工智能平台等基础云计算服务的云服务器,等等。可选的,任一服务器可以位于区块链网络外,也可以位于区块链网络内,对此不作限定。并且,边缘服务器和云端服务器的型号均可以是常见的型号,且边缘服务器和云端服务器可具备指定网速(如100Gbps(每秒1000兆位))的网卡,以提升带宽。
具体的,在图1a所示的数据传输系统中,实现本申请实施例所提出的数据传输方案的大致流程可参见图1b所示,具体如下:
①可基于一个或多个运营商为边缘服务器配置n个公网通信地址,以及为每个云端服务器配置m个公网通信地址;也就是说,边缘服务器的n个公网通信地址以及每个云端服务器的m个公网通信地址,均可由一个或多个运营商提供。其中,n和m均为正整数,且n和m的取值可相同或不同;所谓的公网通信地址是指在公网中所使用的通信地址,其例如可以是公网IP(Internet Protocol,网络之间互联的协议)地址。边缘服务器中的一个公网通信地址和云端服务器中的一个公网通信地址,可用于建立一条运营商链路;基于此,边缘服务器中的传输层可根据边缘服务器的n个公网通信地址以及每个云端服务器的m个公网通信地址,在边缘服务器和每个云端服务器之间均建立n×m条运营商链路。需说明的是,当公网通信地址为公网IP地址时,该运营商链路又可称为IP链路。
②边缘服务器中的链路质量监控层可对传输层中的各条运营商链路进行网络质量的探测。具体的,该链路质量监控层可先搜集传输层中的各条运营商链路的数据传输情况,每条运营商链路的数据传输情况可包括但不限于:每条运营商链路所传输的数据是否丢包、以及当存在丢包时,丢失的数据包的数量等丢包情况,是否存在延迟、以及当存在延迟时,具体的延迟时长等延迟情况。然后,该链路质量监控层可根据搜集到的各条运营商链路的数据传输情况,确定各条运营商链路的网络质量;并基于确定的各条运营商的网络质量,在边缘服务器和每个云端服务器之间维护两个长度为W的链路质量监控数组,边缘服务器和任一云端服务器之间的两个链路质量监控数组可用于存储:边缘服务器和任一云端服务器之间的各条运营商链路的延迟时长,以及各条通信链路的丢包率。示例性的,采用字母a 表示边缘服务器,采用字母x表示任一云端服务器,那么可在边缘服务器和该云端服务器之间维护延迟D i_ax[W](单位秒),以及丢包率L i_ax[W](单位%)这两个链路质量监控数组;其中,0<i_ax<=n×m,i_ax表示边缘服务器a和云端服务器x之间的第i条运营商链路。可选的,链路质量监控层可定时(即每隔时间T,T为预设时长)地对传输层中的各条运营商链路进行网络质量的探测,从而实现每隔时间T根据监控到的网络质量更新所维护的各个链路质量监控数组,进而实现每隔时间T向多路选优层反馈各个链路质量监控数组。当然,在其他实施例中,链路质量监控层也可实时地进行网络质量的探测,或者只执行一次网络质量的探测,对此不作限定。
③边缘服务器中的多路选优层在接收到某信令数据后,可确定该信令数据需到达的云端服务器(后续称为目标云端服务器),并确定边缘服务器和目标云端服务器之间的数据传输策略。其次,可按照该数据传输策略在边缘服务器和目标云端服务器之间进行链路规划,得到多条通信链路,通信链路本质上是运营商链路。然后,可基于链路质量监控层所反馈的各个链路质量监控数组,确定规划得到的各条通信链路的网络质量;并根据网络质量优化目标、结合每条通信链路的网络质量以及每条通信链路的公网通信成本等因素,规划编排出可满足用户需求且成本较低的传输方式。其中,传输方式可以是选择单条通信链路进行数据传输的方式(简称为单链路的传输方式),也可以是选择至少两条通信链路进行数据传输的方式(简称为多链路的传输方式),对此不作限定。且当传输方式是多链路传输方式时,可通过被选择的至少两条通信链路,对信令数据进行多链路的并发传输,或者对信令数据进行多链路的冗余传输;所谓的并发传输是指各通信链路传输的数据不同,冗余传输则是指各通信链路传输的数据相同或部分相同。在确定了数据传输策略和传输方式后,可多路选优层可向传输层下发确定的数据传输策略和传输方式,使得传输层根据该数据传输策略和传输方式,对信令数据进行传输,以将该信令数据传输至目标云端服务器。
基于上述描述可知,本申请实施例所提出的数据传输方案,可通过利用边缘服务器配置的多个运营商的公网通信地址,以及云端服务器配置的多个运营商的公网通信地址来建立多条运营商链路,并启用云端接入以及云端内网互通的功能,动态规划出多条通信链路,从而进一步结合网络质量优化目标,考虑通信链路的网络质量和公网通信成本等因素,灵活地从规划出的多条通信链路中选择出合适的通信链路进行数据传输,这样不仅可实现采用网络质量较优的通信链路进行数据传输,提升数据传输的可靠性,从而提升数据传输效果,还可有效降低用户成本。
基于上述的描述,本申请实施例提出一种云边协同的数据传输方法,该云边协同的数据传输方法可以由上述所提及的边缘服务器执行,边缘服务器可与多个云端服务器进行通信。请参见图2,该云边协同的数据传输方法可包括以下步骤S201-S204:
S201,确定待传输的信令数据需到达的目标云端服务器。
在具体实现中,边缘服务器在获取到待传输的信令数据后,可先解析该信令数据,以得到该信令数据的目的地址,该目的地址可以是多个云端服务器中的任一云端服务器的公网通信地址;然后,边缘服务器可将该信令数据的目的地址所指示的云端服务器,确定为信令数据需到达的目标云端服务器,即目标云端服务器是指信令数据需到达的云端服务器。
其中,信令数据可以是任意类型的数据。例如,该信令数据可以是在边缘服务器或用户设备的运行过程中,所产生的系统运行数据。又如,该信令数据可以是在用户使用客户端的过程中,根据用户针对客户端的操作行为所生成的用户操作数据。此处的客户端可以是音视频客户端、游戏客户端、浏览器客户端等等,那么信令数据可以是根据用户针对音视频客户端的音视频播放操作所生成的用户操作数据,或者根据用户针对游戏客户端的游戏操作所生成的用户操作数据,或者根据用户针对浏览器 客户端的信息选择操作/信息浏览操作/信息搜索操作等所生成的用户操作数据,等等。
上述所提及的游戏客户端所支持的游戏可以是云游戏或者普通游戏,对此不作限定;所谓的云游戏又可称为游戏点播,是一种以云计算为基础的游戏方式,此处的云计算是一种基于互联网的计算方式;在云游戏场景下,游戏并不在玩家用户的游戏客户端,而是在云游戏服务器中运行;由云游戏服务器将游戏场景所涉及的游戏画面编码为视频流,通过网络传输给玩家用户的游戏客户端进行播放;相应的,普通游戏则是指:直接在玩家用户的游戏客户端中运行的游戏。
S202,按照边缘服务器和目标云端服务器之间的数据传输策略,在边缘服务器和目标云端服务器之间进行链路规划,得到多条通信链路。
在本申请实施例中,可在边缘服务器和目标云端服务器之间支持多种传输的策略;具体的,当边缘服务器和目标云端服务器位于同一个地域时,可支持边缘服务器通过公网直接将数据传输至目标云端服务器;当边缘服务器和目标云端服务器位于不同的地域时,仍可支持边缘服务器通过公网直接将数据传输至目标云端服务器,或者支持边缘服务器通过其他的云端服务器,将数据传输至目标云端服务器,例如可支持边缘服务器就近接入一个云端服务器,并由该就近接入的云端服务器将数据传输至目标云端服务器。所谓的“就近接入”是指:根据边缘服务器和各个远端服务器之间的距离,选择距离最近的云端服务器进行接入,以和距离最近的远端服务器进行数据传输;此处所提及的距离可以是指:边缘服务器的地理位置和远端服务器的地理位置之间的地理距离,也可以是指:边缘服务器的公网通信地址和云端服务器的公网通信地址之间的通信距离,对此不作限定。
在具体实现中,边缘服务器可根据实际情况,从上述所提及的多种传输的策略中选择一种策略,作为边缘服务器和目标云端服务器之间的数据传输策略;也就是说,该数据传输策略可用于指示:边缘服务器通过公网将数据传输至目标云端服务器,或者该数据传输策略可用于指示:边缘服务器就近接入云端服务器,并由就近接入的云端服务器通过内网将数据传输至目标云端服务器。当然,此处只是示例性地列举了数据传输策略的两种情况,并非穷举。例如,在其他实施例中,数据传输策略可用于指示:边缘服务器随机接入一个云端服务器,并由随机接入的云端服务器通过内网将数据传输至目标云端服务器;又如,在其他实施例中,数据传输策略可用于指示:边缘服务器根据各个云端服务器的性能,接入性能最好的云端服务器,并由接入的云端服务器通过内网将数据传输至目标云端服务器,等等。
相应的,当数据传输策略用于指示:边缘服务器通过公网将数据传输至目标云端服务器时;步骤S202的具体实施方式可以是:将边缘服务器和所述目标云端服务器之间的多条运营商链路,作为多条通信链路。当数据传输策略用于指示:边缘服务器就近接入云端服务器,并由就近接入的云端服务器通过内网将数据传输至目标云端服务器时,步骤S202的具体实施方式可以是:将边缘服务器和边缘服务器就近接入的云端服务器之间的多条运营商链路,作为多条通信链路。其中,边缘服务器配置有n个公网通信地址,云端服务器被配置有m个公网通信地址,n和m均为正整数;一条运营商链路是采用边缘服务器的一个公网通信地址,和云端服务器的一个公网通信地址建立的。应理解的是,当数据传输策略用于指示:边缘服务器随机接入一个云端服务器,或者接入性能最好的云端服务器等信息时,步骤S202的具体实施方式与就近接入云端服务器时所采用的具体实施方式类似,在此不再赘述。
S203,按照网络质量优化目标,从多条通信链路中选择满足网络质量优化目标的K条通信链路。
具体的,边缘服务器可获取规划得到的多条通信链路中的各条通信链路的网络质量;并按照网络质量优化目标,根据各条通信链路的网络质量,从多条通信链路中选择满足网络质量优化目标的K条通信链路;K为正整数,且K的取值小于或等于规划得到的通信链路的数量。其中,网络质量优化目 标可包括以下至少一项:需达到的丢包率、需达到的延迟时长,等等。相应的,每条通信链路的网络质量可包括以下至少一项:通信链路的丢包率,以及通信链路的延迟时长,等等。
由前述可知,边缘服务器采用单链路的传输方式进行数据传输,也可以采用多链路的传输方式进行数据传输。当边缘服务器采用单链路的传输方式进行数据传输时,边缘服务器需通过步骤S203选择出1条通信链路;即此情况下,可确定K的取值等于1。当边缘服务器采用多链路的传输方式进行数据传输时,边缘服务器需通过步骤S203选择出至少两条通信链路;即此情况下,可确定K的取值大于1。随着K的取值不同,边缘服务器针对步骤S203所采用的具体实施方式也随之不同,具体参见下述描述:
当K的取值等于1时,步骤S203的具体实施方式可以是:首先,可按照网络质量优化目标,根据各条通信链路的网络质量,对多条通信链路进行筛选。具体的,当网络质量优化目标包括需达到的丢包率,通信链路的网络质量包括通信链路的丢包率时,可从多条通信链路中筛选出丢包率小于或等于需达到的丢包率的通信链路。当网络质量优化目标包括需达到的延迟时长,通信链路的网络质量包括通信链路的延迟时长时,可从多条通信链路中筛选出延迟时长小于或等于需达到的延迟时长的通信链路。当网络质量优化目标同时包括需达到的丢包率以及需达到的延迟时长,通信链路的网络质量包括通信链路的丢包率以及通信链路的延迟时长时,可从多条通信链路中筛选出丢包率小于或等于需达到的丢包率,且延迟时长小于或等于需达到的延迟时长的通信链路。
在基于上述三种筛选方式中的任一筛选方式进行通信链路的筛选后,若筛选出的通信链路的数量为1,则可直接选择该筛选出的通信链路作为最终的K条通信链路。若筛选出的通信链路的数量大于1,则可从筛选出的多条通信链路中随机选取一条通信链路作为最终的K条通信链路;或者,也可根据筛选出的各条通信链路的公网通信成本,从筛选出的多条通信链路中选取公网通信成本最低的通信链路,作为最终的K条通信链路。其中,通信链路的公网通信成本包括:公网出口成本和公网入口成本;公网出口成本是指在单位速率下使用公网出口所需的成本,公网入口成本单价是指在单位速率下使用公网入口所需的成本,如公网出口成本和公网入口成本的单位均可以是元/Mbps。那么,公网通信成本最低是指:公网出口成本和公网入口成本的总和最小。任一公网出口成本可以是由边缘服务器从提供该任一公网出口成本对应的公网出口的运营商处获取到的,同理,任一公网入口成本可以是由边缘服务器从提供该任一公网入口成本对应的公网入口的运营商处获取到的。
当K的取值大于1时,边缘服务器可支持多种传输模式,步骤S203的具体实施方式可以是:首先,可按照网络质量优化目标以及每条通信链路的网络质量,从多条通信链路中选择与每种传输模式相适配的H条通信链路,H为大于1的整数;其次,可基于每种传输模式所适配的各条通信链路的公网通信成本,计算每种传输模式的传输成本;然后,可根据每种传输模式的传输成本,从多种传输模式中选择传输成本满足成本条件的传输模式;从被选择的传输模式所适配的H条通信链路中,选取K条通信链路,K≤H。需要说明的是,每种传输模式所适配的通信链路的数量(即H的取值),可相同或不同,对此不作限定。
其中,上述所提及的多种传输模式可包括但不限于:多路并发传输模式,以及多路冗余传输模式。所谓的多路并发传输模式是指:采用至少两条通信链路传输信令数据,每条通信链路传输信令数据的一个数据块,且各个数据块互不相同的模式。所谓的多路冗余传输模式是指:各条通信链路均传输信令数据的模式;或者,多路冗余传输模式是指:采用至少两条通信链路传输信令数据,每条通信链路传输信令数据的一个数据块,且存在至少两个数据块之间存在冗余数据(即重复数据)的模式。
S204,通过K条通信链路传输信令数据。
在具体实施过程中,当K的取值等于1时,即当通过步骤S203从多条通信链路中选择了1条通信链路时,可直接通过被选择的这1条通信链路传输信令数据。当K的取值大于1时,即当通过步骤S203从多条通信链路中选择了至少两条通信链路时,可采用前述所提及的多路冗余传输模式或者多路并发传输模式,通过被选择的至少两条通信链路传输信令数据。
需要说明的是,参见步骤S202的相关描述可知,由于多条通信链路可能是边缘服务器和目标云端服务器之间的运营商链路,也可能是边缘服务器和其他云端服务器(如就近接入的云端服务器)之间的运营商链路;那么,从多条通信链路中选择出的K条通信链路中各条通信链路的终点可能是目标云端服务器,也可能是其他云端服务器(如就近接入的云端服务器)。当K条通信链路中的各条通信链路的终点为目标云端服务器时,则通过K条通信链路传输信令数据,可实现将信令数据直接传输至目标云端服务器,如图3a所示;当通信链路中的各条通信链路的终点为其他云端服务器(如就近接入的云端服务器)时,则通过K条通信链路传输信令数据,可实现将信令数据传输至其他云端服务器(如就近接入的云端服务器),此情况下,需其他云端服务器(如就近接入的云端服务器)将该信令数据转发至目标云端服务器,如图3b所示。
本申请实施例中的边缘服务器在信令数据需到达的目标云端服务器后,可按照边缘服务器和目标远端服务器之间的数据传输策略,在边缘服务器和目标远端服务器之间进行链路规划,并从规划得到的多条通信链路中选择满足网络质量优化目标的K条通信链路,从而通过该K条通信链路传输该信令数据。由于多条通信链路是基于数据传输策略进行动态的链路规划得到的,因此随着数据传输策略的不同,可使得规划得到的多条通信链路不同,这样可有效提升通信链路的灵活性,从而提升数据传输的灵活性;并且,由于K条通信链路满足网络质量优化目标的,可说明这K条通信链路的网络质量是较优的,那么在通过这K条通信链路传输信令数据时,可有效提升数据传输的可靠性,从而提升数据传输效果。
请参见图4,是本申请另一实施例提供的一种云边协同的数据传输方法的流程示意图。该云边协同的数据传输方法可以由上述所提及的边缘服务器执行,边缘服务器可与多个云端服务器进行通信。在本申请实施例中,主要以边缘服务器采用多链路的传输方式进行数据传输为例进行说明;即本申请实施例中的边缘服务器可支持多种传输模式,该多种传输模式可至少包括:多路冗余传输模式以及多路并发传输模式。请参见图4所示,该云边协同的数据传输方法可包括以下步骤S401-S407:
S401,确定待传输的信令数据需到达的目标云端服务器。
S402,按照边缘服务器和目标云端服务器之间的数据传输策略,在边缘服务器和目标云端服务器之间进行链路规划,得到多条通信链路。
在本申请实施例中,主要以数据传输策略用于指示边缘服务器通过公网将数据传输至目标云端服务器;或者用于指示:边缘服务器就近接入云端服务器,并由就近接入的云端服务器通过内网将数据传输至目标云端服务器,这两种情况为例进行阐述。其中,当数据传输策略用于指示:边缘服务器通过公网将数据传输至目标云端服务器时,多条通信链路是指:边缘服务器和目标云端服务器之间的多条运营商链路;当数据传输策略用于指示:所述边缘服务器就近接入云端服务器,并由就近接入的云端服务器通过内网将数据传输至目标云端服务器时,多条通信链路是指:边缘服务器和就近接入的云端服务器之间的多条运营商链路。
其中,数据传输策略的确定方式可以如下:首先,边缘服务器可根据边缘服务器所处的地理位置,以及目标云端服务器所处的地理位置,确定信令数据的通信需求。具体的,若边缘服务器所处的地理位置和目标云端服务器所处的地理位置位于同一个地域,则可确定信令数据的通信需求为同地域的通 信需求;若边缘服务器所处的地理位置和目标云端服务器所处的地理位置未位于同一个地域,则可确定信令数据的通信需求为跨地域的通信需求。在确定了信令数据的通信需求后,若信令数据的通信需求为同地域的通信需求,则可确定数据传输策略为:边缘服务器通过公网将数据传输至目标云端服务器的策略;若信令数据的通信需求为跨地域的通信需求,则边缘服务器可直接确定数据传输策略为:边缘服务器就近接入云端服务器,并由就近接入的云端服务器通过内网将数据传输至目标云端服务器的策略。
或者,若信令数据的通信需求为跨地域的通信需求,则可根据网络优化方案确定数据传输策略。其中,网络优化方案可以是优先考虑传输成本的方案,也可以是优先考虑传输性能的方案;此处所提及的传输性能可通过传输过程的可靠性进行衡量,可靠性和传输性能可成正相关,即可靠性越高,则表明传输性能越好。那么相应的,若网络优化方案为优先考虑传输成本的方案,则由于公网传输所需的传输成本通常小于内网传输所需的成本,因此可确定数据传输策略为:边缘服务器通过公网将数据传输至目标云端服务器的策略。若网络优化方案为优先考虑传输性能的方案,则由于公网传输的传输性能通常低于内网传输所需的成本,因此可确定数据传输策略为:边缘服务器就近接入云端服务器,并由就近接入的云端服务器通过内网将数据传输至目标云端服务器的策略。
其中,上述所提及的网络优化方案可以是用户通过界面操作直接给定的,也可以是在用户存在网络质量优化的用户需求时,根据经验值或者业务需求确定的。具体的,用户通过界面操作直接给定网络优化方案的方式可包括但不限于:用户所使用的终端设备可为用户提供一个需求设置界面,并在该需求设置界面中提供一个方案输入区域,使得用户可在该方案输入区域输入给定的网络优化方案。或者,用户所使用的终端设备可为用户提供一个需求设置界面,并在该需求设置界面中提供一个方案选择区域,该方案选择区域中包括优先考虑传输成本的方案选项以及优先考虑传输性能的方案选项,使得用户可在方案选择区域中选中某个方案选项,以给定网络优化方案。
在用户存在网络质量优化的用户需求时,根据经验值或者业务需求确定网络优化方案的方式可以是:可预先根据经验值或者业务需求,设置优先考虑传输成本的方案或者优先考虑传输性能的方案,作为默认方案。另外,用户所使用的终端设备可在任一用户界面中,为用户提供一个网络质量优化组件,该网络质量优化组件可以理解成是一种“一键优化”的组件。若该网络质量优化组件被触发,则用户所使用的终端设备可将预先设置的默认方案的方案标识发送给边缘服务器,由边缘服务器将该方案标识所指示的方案作为网络优化方案;或者,若该网络质量优化组件被触发,则用户所使用的终端设备可向边缘服务器发送一个组件触发通知,使得边缘服务器在接收到该组件触发通知后,将预先设置的默认方案确定为网络优化方案。
S403,按照网络质量优化目标以及每条通信链路的网络质量,从多条通信链路中选择与每个传输模式相适配的H条通信链路。
其中,H为大于1的整数。
其中,与步骤S402所提及的网络优化方案类似,网络质量优化目标可以是用户通过界面操作直接给定的,也可以是在用户存在网络质量优化的用户需求时,根据经验值或者业务需求确定的。具体的,用户通过界面操作直接给定网络质量优化目标的方式可包括但不限于:用户所使用的终端设备可为用户提供一个需求设置界面,并在该需求设置界面中提供一个延迟时长设置区域以及丢包率设置区域,使得用户可执行如下至少一种操作以实现网络质量优化目标的设置:在该延迟时长设置区域设置需达到的延迟时长,以及在丢包率设备区域设置需达到的丢包率。在用户存在网络质量优化的用户需求时,根据经验值或者业务需求确定网络质量优化目标的方式可以是:可预先根据经验值或者业务需求,设 置默认的丢包率和默认的延迟时长中的至少一项。另外,用户所使用的终端设备可在任一用户界面中,为用户提供一个网络质量优化组件。若该网络质量优化组件被触发,则可将默认的丢包率作为需达到的丢包率,或者将默认的延迟时长作为需达到的延迟时长添加至网络质量优化目标中,或者将默认的丢包率作为需达到的丢包率,以及将默认的延迟时长作为需达到的延迟时长添加至网络质量优化目标中。可见,本申请实施例所提及的网络质量优化目标可包括以下至少一项:需达到的延迟时长,以及需达到的丢包率。
当网络质量优化目标包括:需达到的延迟时长,通信链路的网络质量包括通信链路的延迟时长时;步骤S403的具体实施方式可包括:在任一传输模式下,从多条通信链路中初选出H条通信链路;从初选出的H条通信链路中的各条通信链路的延迟时长中,确定出最小延迟时长;若最小延迟时长小于需达到的延迟时长,则将初选出的H条通信链路,确定为与任一传输模式相适配的H条通信链路;若最小延迟时长不小于需达到的延迟时长,则边缘服务器可重新执行上述步骤,直至确定出与任一传输模式相适配的H条通信链路。
当然,应理解的是,本申请实施例只是示例性地阐述了根据延迟时长选择H条通信链路的具体方式,并非穷举。例如,在其他实施例中,也可对初选出的H条通信链路的延迟时长进行均值计算,得到平均延迟时长;若平均延迟时长小于需达到的延迟时长,则将初选出的H条通信链路,确定为与任一传输模式相适配的H条通信链路。或者,在其他实施例中,还可初选出的H条通信链路中的每条通信链路的延迟时长,是否均小于需达到的延迟时长;若是,则将初选出的H条通信链路,确定为与任一传输模式相适配的H条通信链路。
也就是说,在任一传输模式下,当初选出的H条通信链路被作为与任一传输模式相适配的H条通信链路时,该初选出的H条通信链路需满足如下条件:初选出的H条通信链路的延迟时长中的最小延迟时长,小于需达到的延迟时长;或者,初选出的H条通信链路的平均延迟时长,小于需达到的延迟时长;再或者,初选出的H条通信链路中的每条通信链路的延迟时长,均小于需达到的延迟时长,等等。为便于阐述,后续均以初选出的H条通信链路的延迟时长中的最小延迟时长,小于需达到的延迟时长这一条件为例进行说明。
当网络质量优化目标包括:需达到的丢包率,通信链路的网络质量包括通信链路的丢包率时,若多种传输模式包括多路并发传输模式,则步骤S403的具体实施方式可包括:从多条通信链路中初选出H条通信链路;对初选出的H条通信链路中的各条通信链路的丢包率进行均值运算,得到平均丢包率;s若平均丢包率小于需达到的丢包率,则将初选出的H条通信链路,确定为与多路并发传输模式相适配的H条通信链路;若平均丢包率不小于需达到的丢包率,则边缘服务器可重新执行上述步骤,直至确定出与多路并发传输模式相适配的的H条通信链路。也就是说,在多路并发传输模式下,当初选出的H条通信链路被作为与多路并发传输模式相适配的H条通信链路时,该初选出的H条通信链路需满足如下条件:根据初选出的各条通信链路的丢包率所计算得到的平均丢包率,小于需达到的丢包率。
当网络质量优化目标包括:需达到的丢包率,通信链路的网络质量包括通信链路的丢包率时,若多种传输模式包括多路冗余传输模式,则步骤S403的具体实施方式可包括:从多条通信链路中初选出H条通信链路;采用综合丢包率计算公式,根据初选出的H条通信链路中的各条通信链路的丢包率,计算初选出的H条通信链路的综合丢包率;若初选出的H条通信链路的综合丢包率小于需达到的丢包率,则将初选出的H条通信链路,确定为与多路冗余传输模式相适配的H条通信链路;若综合丢包率不小于需达到的丢包率,则边缘服务器可重新执行上述步骤,直至确定出与多路冗余传输模式相适配的H条通信链路。也就是说,在多路冗余传输模式下,当初选出的H条通信链路被作为与多路冗余传 输模式相适配的H条通信链路时,该初选出的H条通信链路需满足如下条件:根据初选出的各条通信链路的丢包率计算得到的综合丢包率,小于需达到的丢包率。
其中,综合丢包率计算公式如下:
Figure PCTCN2022097602-appb-000001
其中,L i表示初选出的第i条通信链路的丢包率,i∈[1,H];∏表示连乘积的运算。
基于上述描述,当网络质量优化目标同时包括:需达到的延迟时长以及需达到的丢包率,通信链路的网络质量包括通信链路的延迟时长以及通信链路的丢包率时,若在各种传输模式下所初选出H条通信链路被作为与各种传输模式相适配的H条通信链路,则该各种传输模式下所初选出的H条通信通路需满足的条件可分别如下:
①多路并发传输模式:
若数据传输策略用于指示:边缘服务器(采用字母a表示)通过公网将数据传输至目标云端服务器(采用字母b表示),则在多路并发传输模式下,初选出H条通信链路满足下述公式1.1-公式1.3所示的条件时,该初选的H条通信链路可被作为与多路并发传输模式相适配的H条通信链路:
H≤m×n    式1.1
Figure PCTCN2022097602-appb-000002
min D i_ab<D t,0<i≤H     式1.3
若数据传输策略用于指示:边缘服务器(采用字母a表示)就近接入云端服务器(采用字母c表示),并由就近接入的云端服务器通过内网将数据传输至所述目标云端服务器(采用字母b表示),则在多路并发传输模式下,初选出H条通信链路满足下述公式1.4-公式1.6所示的条件时,该初选的H条通信链路可被作为与多路并发传输模式相适配的H条通信链路:
H≤m×n    式1.4
Figure PCTCN2022097602-appb-000003
min D i_ac<D t,0<i≤H    式1.6
在上述公式1.1-公式1.6中,L i_ab表示在边缘服务器和目标云端服务器之间,初选出的H条通信链路中的第i条通信链路的丢包率;L i_ac表示在边缘服务器和就近接入的云端服务器之间,初选出的H条通信链路中的第i条通信链路的丢包率;L t表示需达到的丢包率,D t表示需达到的延迟时长。
②多路冗余传输模式:
若数据传输策略用于指示:边缘服务器(采用字母a表示)通过公网将数据传输至目标云端服务器(采用字母b表示),则在多路冗余传输模式下,初选出H条通信链路满足下述公式2.1-公式2.3所示的条件时,该初选的H条通信链路可被作为与多路冗余传输模式相适配的H条通信链路:
H≤m×n    式2.1
Figure PCTCN2022097602-appb-000004
min D i_ab<D t,0<i≤H   式2.3
若数据传输策略用于指示:边缘服务器(采用字母a表示)就近接入云端服务器(采用字母c表示),并由就近接入的云端服务器通过内网将数据传输至目标云端服务器(采用字母b表示),则在多路冗余传输模式下,初选出H条通信链路满足下述公式2.4-公式2.6所示的条件时,该初选的H条通信链路可被作为与多路冗余传输模式相适配的H条通信链路:
H≤m×n     式2.4
Figure PCTCN2022097602-appb-000005
min D i_ac<D t,0<i≤H    式2.6
在上述公式2.1-公式2.6中,(1-L i_ab)表示在边缘服务器和目标云端服务器之间,初选出的H条通信链路中的第i条通信链路的成功率;(1-L i_ac)表示在边缘服务器和就近接入的云端服务器之间,初选出的H条通信链路中的第i条通信链路的丢包率。
S404,基于每个传输模式所适配的各条通信链路的公网通信成本,计算每个传输模式的传输成本。
在具体实现中,边缘服务器可根据信令数据的传输速率和任一传输模式所适配的每条通信链路的丢包率,计算任一传输模式所适配的每条通信链路成功传输数据分别所需的传输速率。具体的,可先对基准数值(如数值1)和任一传输模式所适配的每条通信链路的丢包率进行差值运算,得到任一传输模式所适配的每条通信链路的成功率;然后,可采用信令数据的传输速率分别除以任一传输模式所适配的每条通信链路的成功率,得到任一传输模式所适配的每条通信链路成功传输数据分别所需的传输速率。
在得到任一传输模式所适配的每条通信链路成功传输数据分别所需的传输速率后,边缘服务器可根据任一传输模式所适配的每条通信链路所需的传输速率,以及任一传输模式所适配的每条通信链路的公网通信成本,计算任一传输模式的公网通信成本。具体的,前述提及了:通信链路的公网通信成本包括:公网出口成本和公网入口成本;那么边缘服务器在计算任一传输模式的公网通信成本时,可根据任一传输模式所适配的每条通信链路所需的传输速率,以及任一传输模式所适配的每条通信链路的公网出口成本,计算任一传输模式的总公网出口成本;具体可计算任一传输模式所适配的每条通信链路所需的传输速率和对应的公网出口成本之间的乘积,得到任一传输模式所适配的每条通信链路的传输公网入口成本,对任一传输模式所适配的各条通信链路的传输公网入口成本进行求和,得到任一传输模式的总公网出口成本。另外,可根据任一传输模式所适配的每条通信链路所需的传输速率,以及任一传输模式所适配的每条通信链路的公网入口成本,计算任一传输模式的总公网入口成本;具体可计算任一传输模式所适配的每条通信链路所需的传输速率和对应的公网入口成本之间的乘积,得到任一传输模式所适配的每条通信链路的传输公网入口成本,对任一传输模式所适配的各条通信链路的传输公网入口成本进行求和,得到任一传输模式的总公网入口成本。然后,可对总公网出口成本和总 公网入口成本进行求和运算,得到任一传输模式的公网通信成本。
在得到任一传输模式的公网通信成本后,边缘服务器可基于任一传输模式的公网通信成本,计算任一传输模式的传输成本。具体的,若数据传输策略用于指示:边缘服务器通过公网将数据传输至目标云端服务器;则基于任一传输模式的公网通信成本,确定任一传输模式的传输成本这一步骤的具体实施方式可以是:将任一传输模式的公网通信成本,确定为任一传输模式的传输成本。若数据传输策略用于指示:边缘服务器就近接入云端服务器,并由就近接入的云端服务器通过内网将数据传输至目标云端服务器;则基于任一传输模式的公网通信成本,确定任一传输模式的传输成本这一步骤的具体实施方式可以是:根据单位速率下的内网通信成本和传输速率,计算总的内网通信成本;具体的,可将单位速率下的内网通信成本和传输速率之间的乘积,作为总的内网通信成本。然后,可对总的内网通信成本和任一传输模式的公网通信成本进行求和运算,得到任一传输模式的传输成本。
需要说明的是,上述所提及的任一传输模式可以是多路冗余传输模式,也可以是多路并发传输模式。基于上述描述可知,随着数据传输策略的不同,各种传输模式的传输成本的计算方式可存在一定的差异。当数据传输策略用于指示:边缘服务器通过公网将数据传输至目标云端服务器时,任一传输模式的传输成本可参见下述公式3.1所示;当数据传输策略用于指示:边缘服务器就近接入云端服务器,并由就近接入的云端服务器通过内网将数据传输至目标云端服务器时,任一传输模式的传输成本可参见下述公式3.2所示。
Figure PCTCN2022097602-appb-000006
Figure PCTCN2022097602-appb-000007
在上述公式3.1和公式3.2中,mincost表示传输成本,S表示信令数据的传输速率(单位Mbps),P i_ab表示边缘服务器和目标云端服务器之间的,任一传输模式所适配的第i条通信链路的公网出口成本(单位元/Mbps),P i_ac表示边缘服务器和就近接入的云端服务器之间的,任一传输模式所适配的第i条通信链路的公网出口成本(单位元/Mbps);Q i_ab表示任一传输成本所适配的第i条通信链路的公网入口成本(单位元/Mbps),B表示内网通信成本(单位元/Mbps)。
S405,根据每个传输模式的传输成本,从多种传输模式中选择传输成本满足成本条件的传输模式。
其中,成本条件可根据业务需求或者经验值设置。例如,成本条件可以是传输成本最小的条件;那么,边缘服务器在执行步骤S405时,便可根据每种传输模式的传输成本,从多种传输模式中选择传输成本最小的传输模式。又如,成本条件可以是传输成本小于成本阈值的条件;那么,边缘服务器在执行步骤S405时,便可根据每种传输模式的传输成本,从多种传输模式中选择传输成本小于成本阈值的任一传输模式。
S406,从被选择的传输模式所适配的H条通信链路中,选取K条通信链路。
其中,K为大于1的整数,且K≤H。
在一种实施方式中,边缘服务器可直接将被选择的传输模式所适配的H条通信链路,确定为K条通信链路。另一种实施方式中,边缘服务器可从被选择的传输模式所适配的H条通信链路中,随机选取K条通信链路。另一种实施方式中,边缘服务器可按照丢包率(或延迟时长)和选取顺序成正比的 关系,根据被选择的传输模式所适配的各条通信链路的丢包率(或延迟时长),从被选择的传输模式所适配的H条通信链路中,选取K条通信链路;所谓的丢包率(或延迟时长)和选取顺序成正比的关系是指:通信链路的丢包率(或延迟时长)越小,则该通信链路越优先被选取的关系。或者,边缘服务器也可综合考虑被选择的传输模式所适配的各条通信链路的延迟时长和丢包率,从被选择的传输模式所适配的H条通信链路中,选取K条通信链路;此情况下被选取的K条通信链路中的各条通信链路需满足:丢包率小于丢包率阈值,且延迟时长小于时长阈值。应理解的是,本申请实施例只是示例性的列举了几种选取K条通信链路的方式,并非穷举。
S407,通过K条通信链路传输信令数据。
在具体实施过程中,边缘服务器可采用被选择的传输模式,根据信令数据为K条通信链路中的每条通信链路分配一个数据包。具体的,若被选择的传输模式为多路冗余传输模式,则边缘服务器可将信令数据作为一个数据包,分别分配给K条通信链路中的每条通信链路;也就是说,此情况下的每套通信链路被分配的数据包本质上均为信令数据。若被选择的传输模式为多路并发传输模式,则边缘服务器可对信令数据进行分块处理,得到K个数据包;将K个数据包分配给K条通信链路,一条通信链路被分配一个数据包;也就是说,此情况下的每条通信链路被分配的数据包本质上均为信令数据的一个数据块,且每条通信链路被分配的数据包互不相同。
在为K条通信链路中的每条通信链路分配一个数据包之后,边缘服务器可通过K条通信链路中的各条通信链路,分别传输相应的数据包。在一种实施方式中,边缘服务器可直接通过K条通信链路中的各条通信链路,分别传输相应的数据包。另一种实施方式中,边缘服务器可确定被选择的传输模式的模式标记,并在K条通信链路中的各条通信链路对应的数据包中添加模式标记;具体的,可在数据包中的包头添加该模式标记。然后,通过K条通信链路中的各条通信链路,分别传输相应的添加了模式标记的数据包;通过添加模式标记,可便于告知数据包的接收端如何处理接收到的数据包,以得到信令数据。示例性的,当被选择的传输模式为多路冗余传输模式时,添加的模式标记可以采用R表示;当被选择的传输模式为多路冗余传输模式时,添加的模式标记可以采用M表示。
本申请实施例中的边缘服务器在信令数据需到达的目标云端服务器后,可实现根据满足用户实际的通信需求的网络优化方案,来确定边缘服务器和目标远端服务器之间的数据传输策略,从而实现根据满足用户需求的数据传输策略,结合边缘服务器和云端服务器之间建立的多条运营商链路,以及云端内网的可靠通信链路,在边缘服务器和目标远端服务器之间进行链路规划,使得规划得到的多条通信链路满足用户需求。针对规划得到的多条通信链路,还可根据满足用户需求的网络质量优化目标,提供多种传输模式,通过求解各种传输模式的传输成本,可实现从多条通信链路中选择出满足网络质量优化目标且传输成本最低的K条通信链路,来传输该信令数据。这样的传输方式,不仅可实现根据用户需求来优化传输方式,还可实现数据传输效果满足用户需求,实现传输成本的最小化,从而提升用户体验和用户粘度;以及还可实现将多链路的传输方式和云端内网可靠的网络传输能力结合,大幅提升云边协同的可靠性,降低协同成本。
基于上述图2和图4所示的数据传输方法实施例的相关描述,本申请实施例还提出了一种云边协同的数据传输方法。该云边协同的数据传输方法可以由当前云端服务器执行,当前云端服务器是指:通过K条通信链路和边缘服务器进行通信的云端服务器。请参见图5所示,该云边协同的数据传输方法可包括以下步骤S501-S503:
S501,接收边缘服务器通过K条通信链路传输的信令数据。
其中,K为正整数。
其中,K条通信链路是按照网络质量优化目标,从多条通信链路中选择的;多条通信链路是按照边缘服务器和目标云端服务器之间的数据传输策略,在边缘服务器和所述目标云端服务器之间进行链路规划得到的,目标云端服务器是指信令数据需到达的云端服务器。
在具体实现中,若K的取值等于1,则表明边缘服务器是通过1条通信链路直接传输信令数据的;那么,此情况下的当前云端服务器在执行步骤S501时,可直接接收边缘服务器通过这1条通信链路传输的信令数据。若K的取值大于1,则表明边缘服务器是通过至少两条通信链路,传输信令数据相关的K个数据包的;那么,此情况下的当前云端服务器在执行步骤S501时,可接收边缘服务器通过K条通信链路传输的K个数据包,一条通信链路传输了一个数据包。
由前述步骤S204的相关描述可知,通过K条通信链路接收信令数据的当前云端服务器可能是目标云端服务器,也可能是边缘服务器就近接入的云端服务器;基于不同情况,当前云端服务器所需执行的数据处理操作是不同的。基于此,当前云端服务器可通过检测信令数据的目的地址是否为当前云端服务器的公网通信地址,来判断信令数据的目的地址是否指向当前云端服务器。若信令数据的目的地址为当前云端服务器的公网通信地址,则表明信令数据的目的地址指向当前云端服务器,即当前云端服务器就是目标云端服务器,此时可执行步骤S502。若信令数据的目的地址不为当前云端服务器的公网通信地址,则表明信令数据的目的地址未指向当前云端服务器,即当前云端服务器不是目标云端服务器,此时可执行步骤S503。
S502,若接收到的信令数据的目的地址指向当前云端服务器,则对接收到的信令数据进行数据处理。
在具体实现中,若当前云端服务器接收到的直接是信令数据,则步骤S502的具体实施方式可以是:若接收到的信令数据的目的地址指向当前云端服务器,则存储该信令数据,或者对该信令数据进行响应处理。例如,信令数据是云游戏中的游戏用户的用户操作数据,则对信令数据进行响应处理可以是:根据该用户操作数据确定待显示的游戏画面,将游戏画面下发至游戏客户端进行显示。
若当前云端服务器接收到的是K个数据包,则步骤S502的具体实施方式可以是:若接收到的信令数据的目的地址指向当前云端服务器,则对K个数据包进行整合,得到信令数据。具体的,当前云端服务器可从K个数据包中解析出模式标记。若解析出的模式标记为多路冗余传输模式的模式标记,则融合K个数据包中的冗余数据,得到信令数据;可选的,若K个数据包存在误码数据包还可基于K个数据包中除误码数据包以外的其他数据包,对该误码数据包进行恢复,以得到信令数据。若解析出的模式标记为多路并发传输模式的模式标记,则对K个数据包进行重排处理,得到信令数据;具体的,每个数据包可具有一个重排序号,可按照重排序号从小到大的顺序对K个数据包进行重排处理,得到信令数据。
S503,若接收到的信令数据未指向当前云端服务器,则通过内网将接收到的信令数据转发至目标云端服务器。
在具体实现中,若当前云端服务器接收到的直接是信令数据,则步骤S503的具体实施方式可以是:若接收到的信令数据未指向当前云端服务器,则基于信令数据的目的地址确定目标云端服务器的内网通信地址,通过确定的内网通信地址,将该信令数据直接转发给目标云端服务器。若当前云端服务器接收到的是K个数据包,则步骤S503的具体实施方式可以是:若接收到的信令数据未指向当前云端服务器,则通过内网将K个数据包转发至目标云端服务器;具体的,可基于信令数据的目的地址确定目标云端服务器的内网通信地址,通过确定的内网通信地址,将该K个数据包转发给目标云端服务器。
本申请实施例中的多条通信链路是基于数据传输策略进行动态的链路规划得到的,因此随着数据 传输策略的不同,可使得规划得到的多条通信链路不同,这样可有效提升通信链路的灵活性,从而提升数据传输的灵活性;并且,由于K条通信链路满足网络质量优化目标的,可说明这K条通信链路的网络质量是较优的,那么在通过这K条通信链路传输信令数据时,可有效提升数据传输的可靠性,从而提升数据传输效果。
在实际应用中,本申请实施例所提出的云边协同的数据传输方法可运用到多种场景中,如云游戏场景,普通游戏场景、音视频播放场景、基于浏览器的信息浏览场景,等等。以云游戏场景为例,在云游戏场景中,为了提升用户体验,除了在边缘部署边缘服务器以外,还可将视频渲染服务器(即用于渲染游戏画面的服务器)部署在距离用户较近的边缘,而游戏服务器(用于提供游戏服务的云端服务器)可能和边缘服务器同地域部署,也可能跨地域部署。虽然,耗时的视频渲染服务器可以就近部署实现,但是根据用户的游戏操作所产生的信令数据仍然需要通过边缘服务器上传到云端服务器(即游戏服务器),以进行数据的同步操作。基于此,为了使得信令数据的数据传输效果达到既能够满足用户需求,又能够优化传输成本的需求,边缘服务器可采用该云边协同的数据传输方案,针对不同的通信需求,综合考虑多链路的通信质量,成本等因素,为信令数据规划编排出优化的传输方式以进行数据传输,提升用户体验。具体的,在云游戏场景中运用该云边协同的数据传输方法的大致过程如下:
首先,用户在通过游戏客户端选择了某一云游戏后,游戏客户端可为用户输出需求设置界面60,该需求设置界面60可支持用户设置网络优化方案以及网络质量优化目标等信息。例如,该需求设置界面60中可包括方案选择区域61以及网络质量输入区域62,如图6a所示;其中,方案选择区域61中可包括优先考虑传输成本的方案选项611以及优先考虑传输性能的方案选项612,网络质量输入区域62中可包括丢包率设置区域621以及延迟时长设置区域622。那么相应的,用户可在方案选择区域61中选中一个目标方案选项,以及分别在丢包率设置区域621中设置需达到的丢包率,在延迟时长设置区域622中设置需达到的延迟时长;然后,用户可通过触发需求设置界面中的需求确认组件63,使得游戏客户端将用户所选择的目标方案选项所指示的方案确定为用户给定的成本优化方案,以及采用用户设置的需达到的丢包率和需达到的延迟时长生成用户给定的网络质量优化目标,并将该成本优化方案和网络质量优化目标上传至边缘服务器,如图6a所示。
在用户的游戏过程中,游戏客户端可根据用户的游戏操作行为生成信令数据,并将信令数据发送至边缘服务器。相应的,当边缘服务器接收到用户所使用的游戏客户端发送的信令数据后,可先确定待传输的信令数据需到达的目标云端服务器(即目标游戏服务器),并根据边缘服务器所处的地理位置,以及目标云端服务器所处的地理位置,确定信令数据的通信需求。然后,可基于信令数据的不同通信需求,采用不同的编排逻辑进行传输方式的编排,具体如下:
若信令数据的通信需求为同地域的通信需求,则边缘服务器可按照边缘服务器通过公网将数据传输至目标云端服务器的策略,将边缘服务器和目标云端服务器之间的多条运营商链路作为多条通信链路,然后基于每条通信链路的公网通信成本和网络质量,综合选择K条通信链路,并采用多路冗余传输模式或者多路并发传输模式,根据信令数据为K条通信链路中的每条通信链路分配一个数据包,然后基于该K条通信链路的各条通信链路分别传输各自的数据包,以实现将该信令数据传输至目标云端服务器。
若信令数据的通信需求为跨地域的通信需求,则边缘服务器可进一步确定用户所选择的成本优化方案。如果用户选择的成本优化方案是优先考虑传输成本的方案,则边缘服务器可优先选择多链路中网络质量排行靠前的多条链路进行公网转发,即边缘服务器可选择边缘服务器通过公网将数据传输至目标云端服务器的策略;如果用户选择的成本优化方案是优先考虑传输性能的方案,则边缘服务器可 将优先选择就近接入云端服务器,并经由就近接入的云端服务器通过内网转发,即边缘服务器可选择边缘服务器就近接入云端服务器,并由就近接入的云端服务器通过内网将数据传输至目标云端服务器的策略。在确定数据传输策略后,边缘服务器可按照该数据传输策略进行链路规划,并基于每条通信链路的公网通信成本和网络质量,进行多路选择以选择出K条通信链路,然后,采用多路冗余传输模式或者多路并发传输模式,根据信令数据为K条通信链路中的每条通信链路分配一个数据包,然后基于该K条通信链路的各条通信链路分别传输各自的数据包,以实现将该信令数据传输至目标云端服务器。
相应的,通过K条通信链路和边缘服务器进行通信的当前云端服务器,可采用图6b所示的操作流程图执行相关操作:
1)通过K条通信链路分别接收相应的数据包,并解析接收到的当前数据包的包头(或称为头部),以得到当前数据包的目的地址。然后,判断当前数据包的目的地址是否指向当前云端服务器;如果当前数据包的目的地址指向当前云端服务器,则则跳转到步骤2);如果当前数据包的目的地址未指向当前云端服务器,则跳转到步骤3)。
2)根据当前数据包的包头中的模式标记执行不同的操作。如果包头中的模式标记为R(即多路冗余传输模式的模式标记),则跳转到步骤a);如果包头中的模式标记为M(即多路并发传输模式的模式标记),则跳转到步骤b);
a)汇总K条通信链路传输的数据包,并且尝试融合K个数据包中的冗余数据,以及对K个数据包中的误码数据包进行恢复,以恢复出信令数据。如果成功恢复信令数据,则向边缘服务器反馈ACK(Acknowledgement,确认)数据包,并跳转到步骤4);如果没有成功恢复信令数据,则向边缘服务器反馈NACK(Negative Acknowledgement,否认确认)数据包,并跳转到步骤1);
b)接收K条通信链路传输的数据包,并尝试对接收的K个数据包进行重排;当K个数据包都完整接收,重排得到信令数据时,跳转到步骤4);当存在丢包时,则向边缘服务器反馈丢失的数据包的标识(ID),并跳转到步骤1)。
3)转发K条通信链路传输的数据包至目标云端服务器。
4)结束流程。
通过在云游戏场景中运用本申请实施例所提出的云边协同的数据传输方案,可实现在云游戏场景中,用户的信令数据如何稳定可靠地传输到云端的游戏服务器,以令游戏服务器的渲染视频流能够及时的回传,从而提升用户游戏体验。另外,还可实现依据用户需求优化传输方式的功能,提升云游戏用户的游戏控制指令传输的可靠性,降低控制时延,提升用户游戏体验。通过将多链路的传输方式和云端内网高可靠的网络传输能力结合,可大幅提升云游戏云边协同的可靠性,降低协同成本。
基于上述图2和图4所示的云边协同的数据处理方法实施例的描述,本申请实施例还公开了一种云边协同的数据处理装置,所述云边协同的数据处理装置可以是运行于边缘服务器中的一个计算机可读指令(包括程序代码)。该云边协同的数据处理装置可以执行图2或图4所示的方法。请参见图7,所述云边协同的数据处理装置可以运行如下单元:
确定单元701,用于确定待传输的信令数据需到达的目标云端服务器;
处理单元702,用于按照边缘服务器和所述目标云端服务器之间的数据传输策略,在所述边缘服务器和所述目标云端服务器之间进行链路规划,得到多条通信链路;
所述处理单元702,还用于按照网络质量优化目标,从所述多条通信链路中选择满足所述网络质量优化目标的K条通信链路,K为正整数;
传输单元703,用于通过所述K条通信链路传输所述信令数据。
在一种实施方式中,所述数据传输策略用于指示:所述边缘服务器通过公网将数据传输至目标云端服务器;相应的,处理单元702在用于按照边缘服务器和所述目标云端服务器之间的数据传输策略,在所述边缘服务器和所述目标云端服务器之间进行链路规划,得到多条通信链路时,可具体用于:
将所述边缘服务器和所述目标云端服务器之间的多条运营商链路,作为多条通信链路。
另一种实施方式中,所述数据传输策略用于指示:所述边缘服务器就近接入云端服务器,并由就近接入的云端服务器通过内网将数据传输至所述目标云端服务器;相应的,处理单元702在用于按照边缘服务器和所述目标云端服务器之间的数据传输策略,在所述边缘服务器和所述目标云端服务器之间进行链路规划,得到多条通信链路时,可具体用于:
将所述边缘服务器和所述边缘服务器就近接入的云端服务器之间的多条运营商链路,作为多条通信链路。
另一种实施方式中,所述边缘服务器配置有n个公网通信地址,云端服务器被配置有m个公网通信地址,n和m均为正整数;
一条运营商链路是采用所述边缘服务器的一个公网通信地址,和云端服务器的一个公网通信地址建立的。
另一种实施方式中,所述边缘服务器支持多种传输模式;相应的,处理单元702在用于按照网络质量优化目标、从所述多条通信链路中选择满足网络质量优化目标的K条通信链路时,可具体用于:
按照网络质量优化目标以及每条通信链路的网络质量,从所述多条通信链路中选择与所述每种传输模式相适配的H条通信链路,H为大于1的整数;
基于所述每种传输模式所适配的各条通信链路的公网通信成本,计算所述每种传输模式的传输成本;
根据所述每种传输模式的传输成本,从所述多种传输模式中选择传输成本满足成本条件的传输模式;所述成本条件包括:传输成本最小的条件,或传输成本小于成本阈值的条件;
从被选择的传输模式所适配的H条通信链路中,选取K条通信链路,K≤H。
另一种实施方式中,所述网络质量优化目标包括:需达到的丢包率,通信链路的网络质量包括通信链路的丢包率;
所述多种传输模式包括多路并发传输模式,处理单元702在用于按照网络质量优化目标以及每条通信链路的网络质量,从所述多条通信链路中选择与每种传输模式相适配的H条通信链路时,可具体用于:
从所述多条通信链路中初选出H条通信链路;
对初选出的H条通信链路中的各条通信链路的丢包率进行均值运算,得到平均丢包率;
若所述平均丢包率小于所述需达到的丢包率,则将所述初选出的H条通信链路,确定为与所述多路并发传输模式相适配的H条通信链路。
另一种实施方式中,所述网络质量优化目标包括:需达到的丢包率,通信链路的网络质量包括通信链路的丢包率;
所述多种传输模式包括多路冗余传输模式,处理单元702在用于按照网络质量优化目标以及每条通信链路的网络质量,从所述多条通信链路中选择与每种传输模式相适配的H条通信链路时,可具体用于:
从所述多条通信链路中初选出H条通信链路;
采用综合丢包率计算公式,根据初选出的H条通信链路中的各条通信链路的丢包率,计算所述初选出的H条通信链路的综合丢包率;
若所述初选出的H条通信链路的综合丢包率小于所述需达到的丢包率,则将所述初选出的H条通信链路,确定为与所述多路冗余传输模式相适配的H条通信链路;
其中,所述综合丢包率计算公式如下:
Figure PCTCN2022097602-appb-000008
其中,L i表示初选出的第i条通信链路的丢包率,i∈[1,H]。
另一种实施方式中,所述网络质量优化目标包括:需达到的延迟时长,通信链路的网络质量包括通信链路的延迟时长;处理单元702在用于按照网络质量优化目标以及每条通信链路的网络质量,从所述多条通信链路中选择与每种传输模式相适配的H条通信链路时,可具体用于:
在任一传输模式下,从所述多条通信链路中初选出H条通信链路;
从初选出的H条通信链路中的各条通信链路的延迟时长中,确定出最小延迟时长;
若所述最小延迟时长小于所述需达到的延迟时长,则将所述初选出的H条通信链路,确定为与所述任一传输模式相适配的H条通信链路。
另一种实施方式中,处理单元702在用于基于所述每种传输模式所适配的各条通信链路的通信成本,计算所述每种传输模式的传输成本时,可具体用于:
根据所述信令数据的传输速率和所述任一传输模式所适配的每条通信链路的丢包率,计算所述任一传输模式所适配的每条通信链路成功传输数据分别所需的传输速率;
根据所述任一传输模式所适配的每条通信链路所需的传输速率,以及所述任一传输模式所适配的每条通信链路的公网通信成本,计算所述任一传输模式的公网通信成本;
基于所述任一传输模式的公网通信成本,计算所述任一传输模式的传输成本。
另一种实施方式中,通信链路的公网通信成本包括:公网出口成本和公网入口成本;所述公网出口成本是指在单位速率下使用公网出口所需的成本,所述公网入口成本单价是指在单位速率下使用公网入口所需的成本;
相应的,处理单元702在用于根据所述任一传输模式所适配的每条通信链路所需的传输速率,以及所述任一传输模式所适配的每条通信链路的公网通信成本,计算所述任一传输模式的公网通信成本时,可具体用于:
根据所述任一传输模式所适配的每条通信链路所需的传输速率,以及所述任一传输模式所适配的每条通信链路的公网出口成本,计算所述任一传输模式的总公网出口成本;
根据所述任一传输模式所适配的每条通信链路所需的传输速率,以及所述任一传输模式所适配的每条通信链路的公网入口成本,计算所述任一传输模式的总公网入口成本;
对所述总公网出口成本和所述总公网入口成本进行求和运算,得到所述任一传输模式的公网通信成本。
另一种实施方式中,所述数据传输策略用于指示:所述边缘服务器通过公网将数据传输至目标云端服务器的策略;
相应的,处理单元702在用于基于所述任一传输模式的公网通信成本,确定所述任一传输模式的传输成本时,可具体用于:将所述任一传输模式的公网通信成本,确定为所述任一传输模式的传输成本。
另一种实施方式中,所述数据传输策略用于指示:所述边缘服务器就近接入云端服务器,并由就近接入的云端服务器通过内网将数据传输至所述目标云端服务器;
相应的,处理单元702在用于基于所述任一传输模式的公网通信成本,确定所述任一传输模式的传输成本时,可具体用于:
根据单位速率下的内网通信成本和所述传输速率,计算总的内网通信成本;
对所述总的内网通信成本和所述任一传输模式的公网通信成本进行求和运算,得到所述任一传输模式的传输成本。
另一种实施方式中,传输单元703在用于通过所述K条通信链路传输所述信令数据时,可具体用于:
采用所述被选择的传输模式,根据所述信令数据为所述K条通信链路中的每条通信链路分配一个数据包;
通过所述K条通信链路中的各条通信链路,分别传输相应的数据包。
另一种实施方式中,所述被选择的传输模式为多路冗余传输模式;相应的,传输单元703在用于采用所述被选择的传输模式,根据所述信令数据为所述K条通信链路中的每条通信链路分配一个数据包时,可具体用于:将所述信令数据作为一个数据包,分别分配给所述K条通信链路中的每条通信链路。
另一种实施方式中,所述被选择的传输模式为多路并发传输模式;相应的,传输单元703在用于采用所述被选择的传输模式,根据所述信令数据为所述K条通信链路中的每条通信链路分配一个数据包时,可具体用于:
对所述信令数据进行分块处理,得到K个数据包;
将所述K个数据包分配给所述K条通信链路,一条通信链路被分配一个数据包。
另一种实施方式中,传输单元703在用于通过所述K条通信链路中的各条通信链路,分别传输相应的数据包时,可具体用于:
确定所述被选择的传输模式的模式标记,并在所述K条通信链路中的各条通信链路对应的数据包中添加所述模式标记;
通过所述K条通信链路中的各条通信链路,分别传输相应的添加了所述模式标记的数据包。
另一种实施方式中,处理单元702还可用于:
根据所述边缘服务器所处的地理位置,以及所述目标云端服务器所处的地理位置,确定所述信令数据的通信需求;
若所述信令数据的通信需求为同地域的通信需求,则确定所述数据传输策略为:所述边缘服务器通过公网将数据传输至目标云端服务器的策略;
若所述信令数据的通信需求为跨地域的通信需求,则根据网络优化方案确定所述数据传输策略。
另一种实施方式中,处理单元702在用于根据网络优化方案确定所述数据传输策略时,可具体用于:
若网络优化方案为优先考虑传输成本的方案,则确定所述数据传输策略为:所述边缘服务器通过公网将数据传输至目标云端服务器的策略;
若网络优化方案为优先考虑传输性能的方案,则确定所述数据传输策略为:所述边缘服务器就近接入云端服务器,并由就近接入的云端服务器通过内网将数据传输至所述目标云端服务器的策略。
根据本申请的另一个实施例,图7所示的云边协同的数据传输装置中的各个单元可以分别或全部 合并为一个或若干个另外的单元来构成,或者其中的某个(些)单元还可以再拆分为功能上更小的多个单元来构成,这可以实现同样的操作,而不影响本申请的实施例的技术效果的实现。上述单元是基于逻辑功能划分的,在实际应用中,一个单元的功能也可以由多个单元来实现,或者多个单元的功能由一个单元实现。在本申请的其它实施例中,基于云边协同的数据传输装置也可以包括其它单元,在实际应用中,这些功能也可以由其它单元协助实现,并且可以由多个单元协作实现。
根据本申请的另一个实施例,可以通过在包括中央处理单元(CPU)、随机存取存储介质(RAM)、只读存储介质(ROM)等处理元件和存储元件的例如计算机的通用计算设备上运行能够执行如图2或图4中所示的相应方法所涉及的各步骤的计算机可读指令(包括程序代码),来构造如图7中所示的云边协同的数据传输装置,以及来实现本申请实施例的游戏处理方法。所述计算机可读指令可以记载于例如计算机可读记录介质上,并通过计算机可读记录介质装载于上述计算设备中,并在其中运行。
本申请实施例中的边缘服务器在信令数据需到达的目标云端服务器后,可按照边缘服务器和目标远端服务器之间的数据传输策略,在边缘服务器和目标远端服务器之间进行链路规划,并从规划得到的多条通信链路中选择满足网络质量优化目标的K条通信链路,从而通过该K条通信链路传输该信令数据。由于多条通信链路是基于数据传输策略进行动态的链路规划得到的,因此随着数据传输策略的不同,可使得规划得到的多条通信链路不同,这样可有效提升通信链路的灵活性,从而提升数据传输的灵活性;并且,由于K条通信链路满足网络质量优化目标的,可说明这K条通信链路的网络质量是较优的,那么在通过这K条通信链路传输信令数据时,可使得有效提升数据传输的可靠性,从而提升数据传输效果。
基于上述图5所示的云边协同的数据处理方法实施例的描述,本申请实施例还公开了一种云边协同的数据处理装置,所述云边协同的数据处理装置可以是运行于当前云端服务器中的一个计算机可读指令(包括程序代码)。该云边协同的数据处理装置可以执行图5所示的方法。请参见图8,所述云边协同的数据处理装置可以运行如下单元:
传输单元801,用于接收边缘服务器通过K条通信链路传输的信令数据,K为正整数;所述K条通信链路是按照网络质量优化目标,从多条通信链路中选择的;所述多条通信链路是按照所述边缘服务器和目标云端服务器之间的数据传输策略,在所述边缘服务器和所述目标云端服务器之间进行链路规划得到的,所述目标云端服务器是指所述信令数据需到达的云端服务器;
处理单元802,用于若接收到的信令数据的目的地址指向当前云端服务器,则对所述接收到的信令数据进行数据处理,所述当前云端服务器是指:通过所述K条通信链路和所述边缘服务器进行通信的云端服务器;
所述传输单元801,还用于若所述接收到的信令数据未指向所述当前云端服务器,则通过内网将所述接收到的信令数据转发至所述目标云端服务器。
在一种实施方式中,传输单元801在用于接收边缘服务器通过K条通信链路传输的信令数据时,可具体用于:接收边缘服务器通过K条通信链路传输的K个数据包;
处理单元802在用于若接收到的信令数据的目的地址指向当前云端服务器,则对所述接收到的信令数据进行数据处理时,可具体用于:若接收到的信令数据的目的地址指向当前云端服务器,则对所述K个数据包进行整合,得到所述信令数据;
所述传输单元801在用于若所述接收到的信令数据未指向所述当前云端服务器,则通过内网将所述接收到的信令数据转发至所述目标云端服务器时,可具体用于:若所述接收到的信令数据未指向所述当前云端服务器,则通过内网将所述K个数据包转发至所述目标云端服务器。
另一种实施方式中,处理单元802在用于对所述K个数据包进行整合,得到所述信令数据时,可具体用于:
从所述K个数据包中解析出模式标记;
若解析出的模式标记为多路冗余传输模式的模式标记,则融合所述K个数据包中的冗余数据,得到所述信令数据;
若解析出的模式标记为多路并发传输模式的模式标记,则对所述K个数据包进行重排处理,得到所述信令数据。
根据本申请的另一个实施例,图8所示的云边协同的数据传输装置中的各个单元可以分别或全部合并为一个或若干个另外的单元来构成,或者其中的某个(些)单元还可以再拆分为功能上更小的多个单元来构成,这可以实现同样的操作,而不影响本申请的实施例的技术效果的实现。上述单元是基于逻辑功能划分的,在实际应用中,一个单元的功能也可以由多个单元来实现,或者多个单元的功能由一个单元实现。在本申请的其它实施例中,基于云边协同的数据传输装置也可以包括其它单元,在实际应用中,这些功能也可以由其它单元协助实现,并且可以由多个单元协作实现。
根据本申请的另一个实施例,可以通过在包括中央处理单元(CPU)、随机存取存储介质(RAM)、只读存储介质(ROM)等处理元件和存储元件的例如计算机的通用计算设备上运行能够执行如图5中所示的相应方法所涉及的各步骤的计算机可读指令(包括程序代码),来构造如图8中所示的云边协同的数据传输装置设备,以及来实现本申请实施例的游戏处理方法。所述计算机可读指令可以记载于例如计算机可读记录介质上,并通过计算机可读记录介质装载于上述计算设备中,并在其中运行。
本申请实施例中的多条通信链路是基于数据传输策略进行动态的链路规划得到的,因此随着数据传输策略的不同,可使得规划得到的多条通信链路不同,这样可有效提升通信链路的灵活性,从而提升数据传输的灵活性;并且,由于K条通信链路满足用户给定的网络质量优化目标的,那么在通过这K条通信链路传输信令数据时,可使得数据传输效果满足用户需求,即可在满足用户需求的情况下,实现信令数据的传输,从而提升用户体验和用户粘度。
基于上述方法实施例以及装置实施例的描述,本申请实施例还提供一种服务器;该服务器可以是前述所提及的边缘服务器,也可以是前述所提及的当前云端服务器。请参见图9,该服务器至少包括处理器、输入接口、输出接口以及计算机存储介质;且该服务器内的处理器、输入接口、输出接口以及计算机存储介质可通过总线或其他方式连接。计算机存储介质可以存储在服务器的存储器中,所述计算机存储介质用于存储计算机可读指令,处理器(或称CPU(Central Processing Unit,中央处理器))是服务器的计算核心以及控制核心,其适于实现一条或多条指令,具体适于加载并执行一条或多条指令从而实现相应方法流程或相应功能。
在一个实施例中,若图9所示的服务器为边缘服务器,则本申请实施例所述的处理器可用于进行一系列的云边协同的数据传输方法。
另一个实施例中,若图9所示的服务器为当前云端服务器,则本申请实施例所述的处理器可用于进行一系列的云边协同的数据传输方法。
本申请实施例还提供了一种计算机存储介质(Memory),所述计算机存储介质是服务器中的记忆设备,用于存放程序和数据。可以理解的是,此处的计算机存储介质既可以包括服务器中的内置存储介质,当然也可以包括服务器所支持的扩展存储介质。计算机存储介质提供存储空间,该存储空间存储了服务器的操作系统。并且,在该存储空间中还存放了适于被处理器加载并执行的一条或多条的指令,这些指令可以是一个或一个以上的计算机可读指令(包括程序代码)。需要说明的是,此处的计算 机存储介质可以是高速RAM存储器,也可以是非不稳定的存储器(non-volatile memory),例如至少一个磁盘存储器;可选的,还可以是至少一个位于远离前述处理器的计算机存储介质。
在一个实施例中,若图9所示的服务器为边缘服务器,则可由处理器加载并执行计算机存储介质中存放的一条或多条指令,以实现上述有关图2或图4所示的云边协同的数据传输方法实施例中的方法的相应步骤。
另一个实施例中,若图9所示的服务器为当前云端服务器,则可由处理器加载并执行计算机存储介质中存放的一条或多条指令,以实现上述有关图5所示的云边协同的数据传输方法实施例中的方法的相应步骤。
需要说明的是,根据本申请的一个方面,还提供了一种计算机程序产品或计算机程序,该计算机程序产品或计算机程序包括计算机指令,该计算机指令存储在计算机可读存储介质中。计算机设备的处理器从计算机可读存储介质读取该计算机指令,处理器执行该计算机指令,使得该计算机设备执行上述图2、图4或图5所示的云边协同的数据传输方法实施例方面的各种可选方式中提供的方法。

Claims (25)

  1. 一种云边协同的数据传输方法,由边缘服务器执行,所述方法包括:
    确定待传输的信令数据需到达的目标云端服务器;
    按照边缘服务器和所述目标云端服务器之间的数据传输策略,在所述边缘服务器和所述目标云端服务器之间进行链路规划,得到多条通信链路;
    按照网络质量优化目标,从所述多条通信链路中选择满足所述网络质量优化目标的K条通信链路,K为正整数;
    通过所述K条通信链路传输所述信令数据。
  2. 如权利要求1所述的方法,其特征在于,所述数据传输策略用于指示:所述边缘服务器通过公网将数据传输至目标云端服务器;
    所述按照边缘服务器和所述目标云端服务器之间的数据传输策略,在所述边缘服务器和所述目标云端服务器之间进行链路规划,得到多条通信链路,包括:
    将所述边缘服务器和所述目标云端服务器之间的多条运营商链路,作为多条通信链路。
  3. 如权利要求1所述的方法,其特征在于,所述数据传输策略用于指示:所述边缘服务器就近接入云端服务器,并由就近接入的云端服务器通过内网将数据传输至所述目标云端服务器;
    所述按照边缘服务器和所述目标云端服务器之间的数据传输策略,在所述边缘服务器和所述目标云端服务器之间进行链路规划,得到多条通信链路,包括:
    将所述边缘服务器和所述边缘服务器就近接入的云端服务器之间的多条运营商链路,作为多条通信链路。
  4. 如权利要求2或3所述的方法,其特征在于,所述边缘服务器配置有n个公网通信地址,云端服务器被配置有m个公网通信地址,n和m均为正整数;
    一条运营商链路是采用所述边缘服务器的一个公网通信地址,和云端服务器的一个公网通信地址建立的。
  5. 如权利要求1所述的方法,其特征在于,所述边缘服务器支持多种传输模式;所述按照网络质量优化目标,从所述多条通信链路中选择满足网络质量优化目标的K条通信链路,包括:
    按照网络质量优化目标以及每条通信链路的网络质量,从所述多条通信链路中选择与每种传输模式相适配的H条通信链路,H为大于1的整数;
    基于所述每种传输模式所适配的各条通信链路的公网通信成本,计算所述每种传输模式的传输成本;
    根据所述每种传输模式的传输成本,从所述多种传输模式中选择传输成本满足成本条件的传输模式;所述成本条件包括:传输成本最小的条件,或传输成本小于成本阈值的条件;
    从被选择的传输模式所适配的H条通信链路中,选取K条通信链路,K≤H。
  6. 如权利要求5所述的方法,其特征在于,所述网络质量优化目标包括:需达到的丢包率,通信链路的网络质量包括通信链路的丢包率;
    所述多种传输模式包括多路并发传输模式,所述按照网络质量优化目标以及每条通信链路的网络质量,从所述多条通信链路中选择与每种传输模式相适配的H条通信链路,包括:
    从所述多条通信链路中初选出H条通信链路;
    对初选出的H条通信链路中的各条通信链路的丢包率进行均值运算,得到平均丢包率;
    若所述平均丢包率小于所述需达到的丢包率,则将所述初选出的H条通信链路,确定为与所述多 路并发传输模式相适配的H条通信链路。
  7. 如权利要求5所述的方法,其特征在于,所述网络质量优化目标包括:需达到的丢包率,通信链路的网络质量包括通信链路的丢包率;
    所述多种传输模式包括多路冗余传输模式,所述按照网络质量优化目标以及每条通信链路的网络质量,从所述多条通信链路中选择与每种传输模式相适配的H条通信链路,包括:
    从所述多条通信链路中初选出H条通信链路;
    采用综合丢包率计算公式,根据初选出的H条通信链路中的各条通信链路的丢包率,计算所述初选出的H条通信链路的综合丢包率;
    若所述初选出的H条通信链路的综合丢包率小于所述需达到的丢包率,则将所述初选出的H条通信链路,确定为与所述多路冗余传输模式相适配的H条通信链路;
    其中,所述综合丢包率计算公式如下:
    Figure PCTCN2022097602-appb-100001
    其中,L i表示初选出的第i条通信链路的丢包率,i∈[1,H]。
  8. 如权利要求5所述的方法,其特征在于,所述网络质量优化目标包括:需达到的延迟时长,通信链路的网络质量包括通信链路的延迟时长;
    所述按照网络质量优化目标以及每条通信链路的网络质量,从所述多条通信链路中选择与每种传输模式相适配的H条通信链路,包括:
    在任一传输模式下,从所述多条通信链路中初选出H条通信链路;
    从初选出的H条通信链路中的各条通信链路的延迟时长中,确定出最小延迟时长;
    若所述最小延迟时长小于所述需达到的延迟时长,则将所述初选出的H条通信链路,确定为与所述任一传输模式相适配的H条通信链路。
  9. 如权利要求5-8任一项所述的方法,其特征在于,所述基于所述每种传输模式所适配的各条通信链路的公网通信成本,计算所述每种传输模式的传输成本,包括:
    根据所述信令数据的传输速率和所述任一传输模式所适配的每条通信链路的丢包率,计算所述任一传输模式所适配的每条通信链路成功传输数据分别所需的传输速率;
    根据所述任一传输模式所适配的每条通信链路成功传输数据分别所需的传输速率,以及所述任一传输模式所适配的每条通信链路的公网通信成本,计算所述任一传输模式的公网通信成本;
    基于所述任一传输模式的公网通信成本,计算所述任一传输模式的传输成本。
  10. 如权利要求9所述的方法,其特征在于,通信链路的公网通信成本包括:公网出口成本和公网入口成本;所述公网出口成本是指在单位速率下使用公网出口所需的成本,所述公网入口成本单价是指在单位速率下使用公网入口所需的成本;
    所述根据所述任一传输模式所适配的每条通信链路成功传输数据分别所需的传输速率,以及所述任一传输模式所适配的每条通信链路的公网通信成本,计算所述任一传输模式的公网通信成本,包括:
    根据所述任一传输模式所适配的每条通信链路成功传输数据分别所需的传输速率,以及所述任一传输模式所适配的每条通信链路的公网出口成本,计算所述任一传输模式的总公网出口成本;
    根据所述任一传输模式所适配的每条通信链路成功传输数据分别所需的传输速率,以及所述任一传输模式所适配的每条通信链路的公网入口成本,计算所述任一传输模式的总公网入口成本;
    对所述总公网出口成本和所述总公网入口成本进行求和运算,得到所述任一传输模式的公网通信成本。
  11. 如权利要求9所述的方法,其特征在于,所述数据传输策略用于指示:所述边缘服务器通过公网将数据传输至目标云端服务器;
    所述基于所述任一传输模式的公网通信成本,确定所述任一传输模式的传输成本,包括:
    将所述任一传输模式的公网通信成本,确定为所述任一传输模式的传输成本。
  12. 如权利要求9所述的方法,其特征在于,所述数据传输策略用于指示:所述边缘服务器就近接入云端服务器,并由就近接入的云端服务器通过内网将数据传输至所述目标云端服务器;
    所述基于所述任一传输模式的公网通信成本,计算所述任一传输模式的传输成本,包括:
    根据单位速率下的内网通信成本和所述传输速率,计算总的内网通信成本;
    对所述总的内网通信成本和所述任一传输模式的公网通信成本进行求和运算,得到所述任一传输模式的传输成本。
  13. 如权利要求5-8任一项所述的方法,其特征在于,所述通过所述K条通信链路传输所述信令数据,包括:
    采用所述被选择的传输模式,根据所述信令数据为所述K条通信链路中的每条通信链路分配一个数据包;
    通过所述K条通信链路中的各条通信链路,分别传输相应的数据包。
  14. 如权利要求13所述的方法,其特征在于,所述被选择的传输模式为多路冗余传输模式,所述采用所述被选择的传输模式,根据所述信令数据为所述K条通信链路中的每条通信链路分配一个数据包,包括:
    将所述信令数据作为一个数据包,分别分配给所述K条通信链路中的每条通信链路。
  15. 如权利要求13所述的方法,其特征在于,所述被选择的传输模式为多路并发传输模式,所述采用所述被选择的传输模式,根据所述信令数据为所述K条通信链路中的每条通信链路分配一个数据包,包括:
    对所述信令数据进行分块处理,得到K个数据包;
    将所述K个数据包分配给所述K条通信链路,一条通信链路被分配一个数据包。
  16. 如权利要求13所述的方法,其特征在于,所述通过所述K条通信链路中的各条通信链路,分别传输相应的数据包,包括:
    确定所述被选择的传输模式的模式标记,并在所述K条通信链路中的各条通信链路对应的数据包中添加所述模式标记;
    通过所述K条通信链路中的各条通信链路,分别传输相应的添加了所述模式标记的数据包。
  17. 如权利要求1所述的方法,其特征在于,所述方法还包括:
    根据所述边缘服务器所处的地理位置,以及所述目标云端服务器所处的地理位置,确定所述信令数据的通信需求;
    若所述信令数据的通信需求为同地域的通信需求,则确定所述数据传输策略为:所述边缘服务器通过公网将数据传输至目标云端服务器的策略;
    若所述信令数据的通信需求为跨地域的通信需求,则根据网络优化方案确定所述数据传输策略。
  18. 如权利要求17所述的方法,其特征在于,所述根据网络优化方案确定所述数据传输策略,包括:
    若网络优化方案为优先考虑传输成本的方案,则确定所述数据传输策略为:所述边缘服务器通过公网将数据传输至目标云端服务器的策略;
    若网络优化方案为优先考虑传输性能的方案,则确定所述数据传输策略为:所述边缘服务器就近接入云端服务器,并由就近接入的云端服务器通过内网将数据传输至所述目标云端服务器的策略。
  19. 一种云边协同的数据传输方法,由云端服务器执行,所述方法包括:
    接收边缘服务器通过K条通信链路传输的信令数据,K为正整数;所述K条通信链路是按照网络质量优化目标,从多条通信链路中选择的;所述多条通信链路是按照所述边缘服务器和目标云端服务器之间的数据传输策略,在所述边缘服务器和所述目标云端服务器之间进行链路规划得到的,所述目标云端服务器是指所述信令数据需到达的云端服务器;
    若接收到的信令数据的目的地址指向当前云端服务器,则对所述接收到的信令数据进行数据处理,所述当前云端服务器是指:通过所述K条通信链路和所述边缘服务器进行通信的云端服务器;
    若所述接收到的信令数据未指向所述当前云端服务器,则通过内网将所述接收到的信令数据转发至所述目标云端服务器。
  20. 如权利要求19所述的方法,其特征在于,所述接收边缘服务器通过K条通信链路传输的信令数据,包括:接收边缘服务器通过K条通信链路传输的K个数据包;
    所述若接收到的信令数据的目的地址指向当前云端服务器,则对所述接收到的信令数据进行数据处理,包括:若接收到的信令数据的目的地址指向当前云端服务器,则对所述K个数据包进行整合,得到所述信令数据;
    所述若所述接收到的信令数据未指向所述当前云端服务器,则通过内网将所述接收到的信令数据转发至所述目标云端服务器,包括:若所述接收到的信令数据未指向所述当前云端服务器,则通过内网将所述K个数据包转发至所述目标云端服务器。
  21. 如权利要求20所述的方法,其特征在于,所述对所述K个数据包进行整合,得到所述信令数据,包括:
    从所述K个数据包中解析出模式标记;
    若解析出的模式标记为多路冗余传输模式的模式标记,则融合所述K个数据包中的冗余数据,得到所述信令数据;
    若解析出的模式标记为多路并发传输模式的模式标记,则对所述K个数据包进行重排处理,得到所述信令数据。
  22. 一种云边协同的数据传输装置,其特征在于,包括:
    确定单元,用于确定待传输的信令数据需到达的目标云端服务器;
    处理单元,用于按照边缘服务器和所述目标云端服务器之间的数据传输策略,在所述边缘服务器和所述目标云端服务器之间进行链路规划,得到多条通信链路;
    所述处理单元,还用于按照网络质量优化目标,从所述多条通信链路中选择满足所述网络质量优化目标的K条通信链路,K为正整数;
    传输单元,用于通过所述K条通信链路传输所述信令数据。
  23. 一种云边协同的数据传输装置,其特征在于,包括:
    传输单元,用于接收边缘服务器通过K条通信链路传输的信令数据,K为正整数;所述K条通信链路是按照网络质量优化目标,从多条通信链路中选择的;所述多条通信链路是按照所述边缘服务器和目标云端服务器之间的数据传输策略,在所述边缘服务器和所述目标云端服务器之间进行链路规划得到的,所述目标云端服务器是指所述信令数据需到达的云端服务器;
    处理单元,用于若接收到的信令数据的目的地址指向当前云端服务器,则对所述接收到的信令数 据进行数据处理,所述当前云端服务器是指:通过所述K条通信链路和所述边缘服务器进行通信的云端服务器;
    所述传输单元,还用于若所述接收到的信令数据未指向所述当前云端服务器,则通过内网将所述接收到的信令数据转发至所述目标云端服务器。
  24. 一种服务器,包括输入接口和输出接口,其特征在于,还包括:
    处理器,适于实现一条或多条计算机可读指令;以及,
    计算机存储介质,所述计算机存储介质存储有一条或多条计算机可读指令,所述一条或多条计算机可读指令适于由所述处理器加载并执行如权利要求1-21任一项所述的云边协同的数据传输方法。
  25. 一种非易失性的计算机可读存储介质,其特征在于,所述非易失性的计算机可读存储介质存储有一条或多条计算机可读指令,所述一条或多条计算机可读指令适于由处理器加载并执行如权利要求1-21任一项所述的云边协同的数据传输方法。
PCT/CN2022/097602 2021-07-13 2022-06-08 云边协同的数据传输方法、服务器及存储介质 WO2023284447A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110793799.4A CN113489643B (zh) 2021-07-13 云边协同的数据传输方法、装置、服务器及存储介质
CN202110793799.4 2021-07-13

Publications (1)

Publication Number Publication Date
WO2023284447A1 true WO2023284447A1 (zh) 2023-01-19

Family

ID=77939095

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/097602 WO2023284447A1 (zh) 2021-07-13 2022-06-08 云边协同的数据传输方法、服务器及存储介质

Country Status (1)

Country Link
WO (1) WO2023284447A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116781697A (zh) * 2023-07-08 2023-09-19 武昌理工学院 基于大数据处理技术的5g信号处理系统及其方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111654541A (zh) * 2020-06-02 2020-09-11 中国联合网络通信集团有限公司 面向边缘计算业务的服务功能链编排方法、系统及编排器
US20200314212A1 (en) * 2019-04-01 2020-10-01 Cloudflare, Inc. Virtual Private Network (VPN) Whose Traffic Is Intelligently Routed
CN111835639A (zh) * 2020-07-06 2020-10-27 杭州网银互联科技股份有限公司 一种基于云计算的sd-wan网络智能链路选择方法
CN112866108A (zh) * 2021-02-01 2021-05-28 迪莲娜(上海)大数据服务有限公司 边缘网关链路择优的方法、系统、介质及云端管理系统
CN113489643A (zh) * 2021-07-13 2021-10-08 腾讯科技(深圳)有限公司 云边协同的数据传输方法、装置、服务器及存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200314212A1 (en) * 2019-04-01 2020-10-01 Cloudflare, Inc. Virtual Private Network (VPN) Whose Traffic Is Intelligently Routed
CN111654541A (zh) * 2020-06-02 2020-09-11 中国联合网络通信集团有限公司 面向边缘计算业务的服务功能链编排方法、系统及编排器
CN111835639A (zh) * 2020-07-06 2020-10-27 杭州网银互联科技股份有限公司 一种基于云计算的sd-wan网络智能链路选择方法
CN112866108A (zh) * 2021-02-01 2021-05-28 迪莲娜(上海)大数据服务有限公司 边缘网关链路择优的方法、系统、介质及云端管理系统
CN113489643A (zh) * 2021-07-13 2021-10-08 腾讯科技(深圳)有限公司 云边协同的数据传输方法、装置、服务器及存储介质

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116781697A (zh) * 2023-07-08 2023-09-19 武昌理工学院 基于大数据处理技术的5g信号处理系统及其方法

Also Published As

Publication number Publication date
CN113489643A (zh) 2021-10-08

Similar Documents

Publication Publication Date Title
US7710869B1 (en) Packet routing to reduce susceptibility to disturbances
US7890656B2 (en) Transmission system, delivery path controller, load information collecting device, and delivery path controlling method
US9088511B2 (en) Multi-hop error recovery
US9338096B2 (en) Multicast tree packing for multi-party video conferencing under SDN environment
US8000239B2 (en) Method and system for bandwidth allocation using router feedback
US7512705B2 (en) Truncating data units
CN113037440B (zh) 数据重传处理方法、装置、计算机设备和存储介质
EP2629466B1 (en) Method, device and system for forwarding data in communication system
CN102726031A (zh) 内容处理方法、装置和系统
CN110191065B (zh) 基于软件定义网络的高性能负载均衡系统与方法
EP1411678A1 (en) Method and system for content-oriented routing of packets in a storage-embedded network
CN104539531A (zh) 数据传输方法及装置
CN103548307A (zh) 通过计算机网络传输数据的方法和系统
CN107770085B (zh) 一种网络负载均衡方法、设备及系统
Kim et al. Improved content management for information-centric networking in SDN-based wireless mesh network
WO2023284447A1 (zh) 云边协同的数据传输方法、服务器及存储介质
CN113472646A (zh) 一种数据传输方法、节点、网络管理器及系统
CN105634968A (zh) 用于控制数据流量的传输的装置及方法
Ye et al. PTP: Path-specified transport protocol for concurrent multipath transmission in named data networks
CN116708598A (zh) 用于实时网络传输的系统及方法
US20080137654A1 (en) Method of managing signaling message in path-based signaled paths to mpls-enabled core network
CN109104376A (zh) 一种数据的转发方法、装置、堆叠设备和计算机可读介质
EP1835666B1 (en) Communication device, routing method, and program
CN111464448B (zh) 一种数据传输方法及装置
CN101364940B (zh) 一种路由器内部通信方法及路由设备

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22841087

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE